{ "meta": { "source": "atlas_v13_current.html", "snapshot": "2026-10-18", "site_version": "13.2", "count": 51, "note": "Materialized view of the current audited graph; complete node fields retained. Source inspection and publication status are recorded per item. The builder performs no new empirical validation." }, "records": [ { "id": "O001", "label": "The Code Nobody Reads", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:Engineering structure" ], "actual": [ "C:C08" ], "mapped": [ "C:C08" ], "printed": true, "summary": "When human line-by-line reading no longer scales, independently justified checks and accountable approval must replace its assurance role.", "limitation": "Predictions and vendor/company statistics are reported second-hand; do not treat a risk-tier policy as externally validated.", "url": "https://addyo.substack.com/p/the-code-nobody-reads", "authors": "Addy Osmani", "year": "2026-09-28", "attributes": { "label": "The Code Nobody Reads", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-code-nobody-reads", "source_url": "https://addyo.substack.com/p/the-code-nobody-reads", "url": "https://addyo.substack.com/p/the-code-nobody-reads", "publication_date": "2026-09-28", "contribution": "When human line-by-line reading no longer scales, independently justified checks and accountable approval must replace its assurance role.", "limitation": "Predictions and vendor/company statistics are reported second-hand; do not treat a risk-tier policy as externally validated.", "manual_use": "Use explicit risk tiers and an evidence dossier rather than assuming every line is read; Keep independent tests and a named human acceptance authority; Protect mentoring, explainability and reviewer capacity.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani revises his earlier advice to inspect every generated line: routine low-risk changes may receive automated first-pass review, while high-blast-radius changes retain deeper scrutiny. He argues that review also transmitted knowledge and that engineering teams must retain both understanding and authority to refuse a release.", "in_short": "When human line-by-line reading no longer scales, independently justified checks and accountable approval must replace its assurance role.", "main_findings": "Use explicit risk tiers and an evidence dossier rather than assuming every line is read; Keep independent tests and a named human acceptance authority; Protect mentoring, explainability and reviewer capacity.", "bibliography_year": "2026", "section_id": "8.8", "chapter_id": "C08", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W093;W100;W111", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "The page includes a Teleport-sponsored passage, alongside quoted vendor/self-reported statistics. Its risk-calibrated review argument is practitioner commentary, not a controlled intervention; validate independent primary metrics before reusing any numbers.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "The page includes a Teleport-sponsored passage, alongside quoted vendor/self-reported statistics. Its risk-calibrated review argument is practitioner commentary, not a controlled intervention; validate independent primary metrics before reusing any numbers.", "editorial_revision_detail": "Secondhand empirical or vendor statistics must not be mistaken for independently replicated studies.; Disclose the explicitly marked sponsorship and vendor telemetry limitations in the September 2026 article.", "prior_record_id": "AOS001", "stable_bibliography_id": "O001", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "8.8", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C08\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Review capacity and maintainers", "K:Independent verification and oracles", "K:Ownership and knowledge transfer" ], "volumes": [ "V06" ] }, { "id": "O002", "label": "Brownfield Agentic Engineering", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:System knowledge & memory", "T:Engineering structure" ], "actual": [ "C:C09" ], "mapped": [ "C:C09" ], "printed": true, "summary": "Reliable agents in mature codebases need recovered tacit constraints and regression evidence before modifying code.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/brownfield-agentic-engineering", "authors": "Addy Osmani", "year": "2026-09-14", "attributes": { "label": "Brownfield Agentic Engineering", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "System knowledge & memory", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/brownfield-agentic-engineering", "source_url": "https://addyo.substack.com/p/brownfield-agentic-engineering", "url": "https://addyo.substack.com/p/brownfield-agentic-engineering", "publication_date": "2026-09-14", "contribution": "Reliable agents in mature codebases need recovered tacit constraints and regression evidence before modifying code.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Inventory external invariants and dependencies before delegation; Make brownfield changes incremental and reversible; Verify old behavior alongside new acceptance criteria.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay stresses that a repository rarely captures the full behavior of a legacy service: contracts, migrations, user journeys and institutional knowledge remain external. Small reversible changes, context gathering and reproducible tests are offered as controls against locally successful but architecturally wrong agent patches.", "in_short": "Reliable agents in mature codebases need recovered tacit constraints and regression evidence before modifying code.", "main_findings": "Inventory external invariants and dependencies before delegation; Make brownfield changes incremental and reversible; Verify old behavior alongside new acceptance criteria.", "bibliography_year": "2026", "section_id": "9.1", "chapter_id": "C09", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W154;W155;W171", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS002", "stable_bibliography_id": "O002", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "9.1", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C09\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:System understanding and rationale", "K:Specifications and governance", "K:Independent verification and oracles" ], "volumes": [ "V06" ] }, { "id": "O003", "label": "Agentic Skill Decay", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory" ], "actual": [ "C:C10" ], "mapped": [ "C:C10" ], "printed": true, "summary": "Developers must deliberately practice reasoning, debugging and code comprehension when agents perform most implementation work.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/agentic-skill-decay", "authors": "Addy Osmani", "year": "2026-08-31", "attributes": { "label": "Agentic Skill Decay", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/agentic-skill-decay", "source_url": "https://addyo.substack.com/p/agentic-skill-decay", "url": "https://addyo.substack.com/p/agentic-skill-decay", "publication_date": "2026-08-31", "contribution": "Developers must deliberately practice reasoning, debugging and code comprehension when agents perform most implementation work.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Preserve manual practice and counterfactual reasoning exercises; Measure retained comprehension rather than prompt throughput; Avoid confusing agent task success with human competence.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani describes how automatic completion can remove the repetitions that once built expertise. He proposes predicting failures before asking an agent, explaining diffs, reconstructing decisions and periodically solving problems manually so proficiency does not disappear behind apparent task completion.", "in_short": "Developers must deliberately practice reasoning, debugging and code comprehension when agents perform most implementation work.", "main_findings": "Preserve manual practice and counterfactual reasoning exercises; Measure retained comprehension rather than prompt throughput; Avoid confusing agent task success with human competence.", "bibliography_year": "2026", "section_id": "10.8", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W117;W118;W121", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "Corrected canonical H1 after verifying public page; earlier erroneous correction was based on browser title.", "prior_record_id": "AOS003", "stable_bibliography_id": "O003", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "title_correction_note": "Publisher page H1: Agentic Skill Decay; browser/SEO metadata: Mastery Still Comes From Doing the Reps; canonicalize on page H1.", "manuscript_section_id": "10.8", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "alternate_browser_title": "Mastery Still Comes From Doing the Reps", "ndse2_actual_chapters_json": "[\"C:C10\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Cognitive offloading and skill debt", "K:Skills formation and teaching", "K:Human cognitive load" ], "volumes": [ "V06" ] }, { "id": "O004", "label": "Audit your Agent files", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:System knowledge & memory", "T:Code quality & security" ], "actual": [ "C:C09" ], "mapped": [ "C:C09" ], "printed": true, "summary": "Agent configuration and memory are perishable operational assets that should be periodically revalidated.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/audit-your-agent-files", "authors": "Addy Osmani", "year": "2026-08-27", "attributes": { "label": "Audit your Agent files", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "System knowledge & memory", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/audit-your-agent-files", "source_url": "https://addyo.substack.com/p/audit-your-agent-files", "url": "https://addyo.substack.com/p/audit-your-agent-files", "publication_date": "2026-08-27", "contribution": "Agent configuration and memory are perishable operational assets that should be periodically revalidated.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Record an owner and expiry policy for agent instructions; Test whether each instruction changes verified behavior; Audit tool privileges and memory freshness independently.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Instructions in AGENTS.md, CLAUDE.md, skills and durable memory can become outdated as tools and repositories change. Osmani recommends auditing usefulness, trimming redundant directives, testing outcomes and separating genuinely persistent facts from temporary guidance.", "in_short": "Agent configuration and memory are perishable operational assets that should be periodically revalidated.", "main_findings": "Record an owner and expiry policy for agent instructions; Test whether each instruction changes verified behavior; Audit tool privileges and memory freshness independently.", "bibliography_year": "2026", "section_id": "9.6", "chapter_id": "C09", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W053;W157;W159", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS004", "stable_bibliography_id": "O004", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "9.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C09\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Agent context and persistent memory", "K:Dependencies and agent supply chain", "K:Specifications and governance" ], "volumes": [ "V06" ] }, { "id": "O005", "label": "Human judgment doesn't leave the software factory. It relocates.", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "Agentic factories relocate human responsibility toward intent, quality criteria and authorization rather than eliminating it.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/human-judgment-doesnt-leave-the-software", "authors": "Addy Osmani", "year": "2026-08-21", "attributes": { "label": "Human judgment doesn't leave the software factory. It relocates.", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/human-judgment-doesnt-leave-the-software", "source_url": "https://addyo.substack.com/p/human-judgment-doesnt-leave-the-software", "url": "https://addyo.substack.com/p/human-judgment-doesnt-leave-the-software", "publication_date": "2026-08-21", "contribution": "Agentic factories relocate human responsibility toward intent, quality criteria and authorization rather than eliminating it.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Put human judgment at the intention and acceptance boundaries; Automate early regression and security checks; Escalate design ambiguity and irreversible operations to authorized owners.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The article compares human-supervised and largely unattended factories and asks who defines product intent, architectural trade-offs and unacceptable regressions. It advocates persistent checks, managed risk boundaries and explicit ownership even when many implementation steps run without a person present.", "in_short": "Agentic factories relocate human responsibility toward intent, quality criteria and authorization rather than eliminating it.", "main_findings": "Put human judgment at the intention and acceptance boundaries; Automate early regression and security checks; Escalate design ambiguity and irreversible operations to authorized owners.", "bibliography_year": "2026", "section_id": "11.6", "chapter_id": "C11", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W147;W149;W093", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS005", "stable_bibliography_id": "O005", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Human-agent collaboration", "K:Specifications and governance", "K:Independent verification and oracles" ], "volumes": [] }, { "id": "O006", "label": "Practical Loop Engineering", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [ "C:C03" ], "mapped": [ "C:C03" ], "printed": true, "summary": "Effective autonomous loops begin with bounded goals, observable checks and explicit stopping rules.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/practical-loop-engineering", "authors": "Addy Osmani", "year": "2026-08-14", "attributes": { "label": "Practical Loop Engineering", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/practical-loop-engineering", "source_url": "https://addyo.substack.com/p/practical-loop-engineering", "url": "https://addyo.substack.com/p/practical-loop-engineering", "publication_date": "2026-08-14", "contribution": "Effective autonomous loops begin with bounded goals, observable checks and explicit stopping rules.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "State an independent success predicate before granting retries; Limit cost, duration, permissions and edit scope; Escalate on ambiguity or verification exhaustion.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Drawing on concurrent agent use, Osmani separates tasks safe for fully bounded delegation from tasks that require close review. He describes iterative act-test-revise operation while emphasizing constraints, definition of done and final human judgment where risk is material.", "in_short": "Effective autonomous loops begin with bounded goals, observable checks and explicit stopping rules.", "main_findings": "State an independent success predicate before granting retries; Limit cost, duration, permissions and edit scope; Escalate on ambiguity or verification exhaustion.", "bibliography_year": "2026", "section_id": "3.6", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W095;W191;W152", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS006", "stable_bibliography_id": "O006", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "3.8", "manuscript_section_id": "3.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C03\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Specifications and governance" ], "volumes": [ "V06" ] }, { "id": "O007", "label": "Agentic Code Quality", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Code quality & security", "T:Verification & review" ], "actual": [], "mapped": [ "C:C05" ], "printed": false, "summary": "Quality assurance for generated code increasingly depends on repeatable harness checks, not only manual inspection or agent opinion.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/agentic-code-quality", "authors": "Addy Osmani", "year": "2026-08-08", "attributes": { "label": "Agentic Code Quality", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Code quality & security", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/agentic-code-quality", "source_url": "https://addyo.substack.com/p/agentic-code-quality", "url": "https://addyo.substack.com/p/agentic-code-quality", "publication_date": "2026-08-08", "contribution": "Quality assurance for generated code increasingly depends on repeatable harness checks, not only manual inspection or agent opinion.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Move quality requirements into repeatable repository checks and agent harness gates; Track defects, architectural fit and post-merge maintenance—not merely model review approval; Retain risk-sensitive human review because multiple models can share error modes.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani notes the scalability limit of inspecting high-volume agent diffs and the instability of repeated model-review judgments. He favors consistent repository rules, static checks, tests and explicit quality gates surrounding generation, while keeping focused human scrutiny for risky changes.", "in_short": "Quality assurance for generated code increasingly depends on repeatable harness checks, not only manual inspection or agent opinion.", "main_findings": "Move quality requirements into repeatable repository checks and agent harness gates; Track defects, architectural fit and post-merge maintenance—not merely model review approval; Retain risk-sensitive human review because multiple models can share error modes.", "bibliography_year": "2026", "section_id": "5.1", "chapter_id": "C05", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W054;W055;W079", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "Repeatable quality gates belong under definitions and operational measures of code quality, not primarily multi-project quality tracing.; Put practical measured quality criteria ahead of generic model-agreement claims.", "prior_record_id": "AOS007", "stable_bibliography_id": "O007", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Technical debt and complexity", "K:Independent verification and oracles", "K:Review capacity and maintainers" ], "volumes": [] }, { "id": "O008", "label": "Software Factories, Light and Dark", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Economics" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "A software factory is many governed loops; its level of human involvement changes both risk and total operating cost.", "limitation": "A conceptual taxonomy and projections, not a controlled comparison of factory operating models.", "url": "https://addyo.substack.com/p/software-factories-light-and-dark", "authors": "Addy Osmani", "year": "2026-07-22", "attributes": { "label": "Software Factories, Light and Dark", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/software-factories-light-and-dark", "source_url": "https://addyo.substack.com/p/software-factories-light-and-dark", "url": "https://addyo.substack.com/p/software-factories-light-and-dark", "publication_date": "2026-07-22", "contribution": "A software factory is many governed loops; its level of human involvement changes both risk and total operating cost.", "limitation": "A conceptual taxonomy and projections, not a controlled comparison of factory operating models.", "manual_use": "Separate an autonomous loop from the governance system supervising many loops; Measure cost of failures and rework, not just generation speed; Choose review intensity by reversibility and blast radius.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay contrasts \"light\" factories that retain human interpretation with \"dark\" factories that minimize review, exposing the trade-off between throughput and lost comprehension. Osmani argues that harness capability and measurable verification should determine safe autonomy, not the attraction of unattended execution.", "in_short": "A software factory is many governed loops; its level of human involvement changes both risk and total operating cost.", "main_findings": "Separate an autonomous loop from the governance system supervising many loops; Measure cost of failures and rework, not just generation speed; Choose review intensity by reversibility and blast radius.", "bibliography_year": "2026", "section_id": "11.9", "chapter_id": "C11", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W027;W147;W152", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "A conceptual taxonomy and projections, not a controlled comparison of factory operating models.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "A conceptual taxonomy and projections, not a controlled comparison of factory operating models.", "editorial_revision_detail": "", "prior_record_id": "AOS008", "stable_bibliography_id": "O008", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Human-agent collaboration", "K:Long-running agent execution", "K:Economic costs and return" ], "volumes": [] }, { "id": "O009", "label": "Own the Outer Loop", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [ "C:C11" ], "mapped": [ "C:C11" ], "printed": true, "summary": "Engineers can delegate the inner execution loop while retaining ownership of quality criteria, verdicts and consequences.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/own-the-outer-loop", "authors": "Addy Osmani", "year": "2026-07-09", "attributes": { "label": "Own the Outer Loop", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/own-the-outer-loop", "source_url": "https://addyo.substack.com/p/own-the-outer-loop", "url": "https://addyo.substack.com/p/own-the-outer-loop", "publication_date": "2026-07-09", "contribution": "Engineers can delegate the inner execution loop while retaining ownership of quality criteria, verdicts and consequences.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Define quality signals before execution; Separate the checker from the final acceptance authority; Require a named owner with authority to stop or roll back.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani distinguishes producing code and testing artifacts from accepting the evidence those checks produce. He locates professional accountability in the surrounding outer loop: intent, checks, risk appetite, provenance, approval and action if a released change fails.", "in_short": "Engineers can delegate the inner execution loop while retaining ownership of quality criteria, verdicts and consequences.", "main_findings": "Define quality signals before execution; Separate the checker from the final acceptance authority; Require a named owner with authority to stop or roll back.", "bibliography_year": "2026", "section_id": "11.6", "chapter_id": "C11", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W095;W100;W111", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS009", "stable_bibliography_id": "O009", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "11.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C11\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Ownership and knowledge transfer", "K:Independent verification and oracles", "K:Human-agent collaboration" ], "volumes": [ "V06" ] }, { "id": "O010", "label": "Agentic Autonomy Levels", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure" ], "actual": [ "C:C03" ], "mapped": [ "C:C03" ], "printed": true, "summary": "Osmani separates autonomy into agency and orchestration axes, describing six operating levels from suggestions to management-by-exception.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/agentic-autonomy-levels", "authors": "Addy Osmani", "year": "2026-07-03", "attributes": { "label": "Agentic Autonomy Levels", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/agentic-autonomy-levels", "source_url": "https://addyo.substack.com/p/agentic-autonomy-levels", "url": "https://addyo.substack.com/p/agentic-autonomy-levels", "publication_date": "2026-07-03", "contribution": "Osmani separates autonomy into agency and orchestration axes, describing six operating levels from suggestions to management-by-exception.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Grade individual-agent initiative separately from multi-agent orchestration; Use the six levels as a contextual operational taxonomy, not a validated maturity index; Set an explicit risk and verification ceiling for delegated work.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani argues that a one-dimensional autonomy ladder hides the difference between what one agent may do and the ability to coordinate many agents. He proposes two axes—agency and orchestration—and six practical levels: assist, supervised action, scoped task delegation, goal-driven autonomy, parallel delegation, and managed-by-exception orchestration. Risk, reversibility, independent checks and escalation limit the appropriate level for any task.", "in_short": "Osmani separates autonomy into agency and orchestration axes, describing six operating levels from suggestions to management-by-exception.", "main_findings": "Grade individual-agent initiative separately from multi-agent orchestration; Use the six levels as a contextual operational taxonomy, not a validated maturity index; Set an explicit risk and verification ceiling for delegated work.", "bibliography_year": "2026", "section_id": "3.3", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W191;W152;W095", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "editorial_revision_detail": "W192 is a data-resource study, not independent validation of a two-axis autonomy framework; W152 is a better research-roadmap comparator.; The original essay explicitly introduces two dimensions (agency and orchestration) and six levels; the earlier one-dimensional paraphrase dropped the framework.; Represent the genuine two-axis framework and its six levels rather than reducing it to a generic risk-tier recommendation.; Capture the two axes, scope of six levels, and lack of empirical validation.; Agency vs orchestration is best represented through execution, human-agent delegation and escalation.; Do not promote conceptual taxonomy or market hypotheses into established research outcomes.", "prior_record_id": "AOS010", "stable_bibliography_id": "O010", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "3.3", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C03\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Long-running agent execution", "K:Human-agent collaboration", "K:Agent failure and escalation" ], "volumes": [ "V06" ] }, { "id": "O011", "label": "Agentic Code Review", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C08" ], "printed": false, "summary": "AI-generated code moves the bottleneck from writing to justified acceptance, making review strategy a core engineering skill.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/agentic-code-review", "authors": "Addy Osmani", "year": "2026-06-16", "attributes": { "label": "Agentic Code Review", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/agentic-code-review", "source_url": "https://addyo.substack.com/p/agentic-code-review", "url": "https://addyo.substack.com/p/agentic-code-review", "publication_date": "2026-06-16", "contribution": "AI-generated code moves the bottleneck from writing to justified acceptance, making review strategy a core engineering skill.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Allocate human attention by risk rather than diff size; Audit automated reviewer false positives and misses; Preserve review as a team knowledge-transfer mechanism.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani contrasts solo experiments and long-lived team repositories, arguing that risk, context, integration and shared learning demand different review methods. Review must consider correctness evidence, comprehension and how design decisions survive the handoff.", "in_short": "AI-generated code moves the bottleneck from writing to justified acceptance, making review strategy a core engineering skill.", "main_findings": "Allocate human attention by risk rather than diff size; Audit automated reviewer false positives and misses; Preserve review as a team knowledge-transfer mechanism.", "bibliography_year": "2026", "section_id": "8.8", "chapter_id": "C08", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W069;W093;W098", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "editorial_revision_detail": "Secondhand empirical or vendor statistics must not be mistaken for independently replicated studies.", "prior_record_id": "AOS011", "stable_bibliography_id": "O011", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Review capacity and maintainers", "K:AI as reviewer", "K:Ownership and knowledge transfer" ], "volumes": [] }, { "id": "O012", "label": "Loop Engineering", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [ "C:C03" ], "mapped": [ "C:C03" ], "printed": true, "summary": "Osmani identifies five interoperable loop-building primitives—scheduling, worktrees, skills, connectors and subagents—plus durable external memory.", "limitation": "A practitioner engineering pattern; the article does not establish comparative loop reliability by controlled experiment.", "url": "https://addyo.substack.com/p/loop-engineering", "authors": "Addy Osmani", "year": "2026-06-08", "attributes": { "label": "Loop Engineering", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/loop-engineering", "source_url": "https://addyo.substack.com/p/loop-engineering", "url": "https://addyo.substack.com/p/loop-engineering", "publication_date": "2026-06-08", "contribution": "Osmani identifies five interoperable loop-building primitives—scheduling, worktrees, skills, connectors and subagents—plus durable external memory.", "limitation": "A practitioner engineering pattern; the article does not establish comparative loop reliability by controlled experiment.", "manual_use": "Schedule discovery and triage but bound side effects; Isolate parallel work in worktrees; share intentions via tested skills/connectors and durable state; Separate the maker from the checker, measure token usage, and escalate when independent checks fail.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Loop engineering means configuring a system that discovers work, dispatches agents, checks outputs, records progress and repeats rather than manually prompting each step. Osmani identifies five recurring primitives: automations/scheduling, isolated Git worktrees, reusable skills, plugins and connectors, and specialized subagents. A separate persistent memory or task board is the sixth requirement for continuity. He also warns about verification, comprehension debt and unpredictable token costs.", "in_short": "Osmani identifies five interoperable loop-building primitives—scheduling, worktrees, skills, connectors and subagents—plus durable external memory.", "main_findings": "Schedule discovery and triage but bound side effects; Isolate parallel work in worktrees; share intentions via tested skills/connectors and durable state; Separate the maker from the checker, measure token usage, and escalate when independent checks fail.", "bibliography_year": "2026", "section_id": "3.6", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W075;W095;W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "A practitioner engineering pattern; the article does not establish comparative loop reliability by controlled experiment. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "A practitioner engineering pattern; the article does not establish comparative loop reliability by controlled experiment. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "editorial_revision_detail": "Article explicitly names five pieces and a sixth durable-memory requirement.; Recover the specific engineering primitives and preserved-state mechanism in Osmani’s source article.; Provide actionable primitives and their limits rather than an abstract autonomous-loop description.; Do not promote conceptual taxonomy or market hypotheses into established research outcomes.", "prior_record_id": "AOS012", "stable_bibliography_id": "O012", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "3.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C03\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Economic costs and return" ], "volumes": [ "V06", "V05" ] }, { "id": "O013", "label": "Long-running Agents", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do" ], "actual": [ "C:C09" ], "mapped": [ "C:C09" ], "printed": true, "summary": "Agents operating across sessions require durable task state, resumability, recovery and verifiable progress.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/long-running-agents", "authors": "Addy Osmani", "year": "2026-04-30", "attributes": { "label": "Long-running Agents", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "System knowledge & memory", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/long-running-agents", "source_url": "https://addyo.substack.com/p/long-running-agents", "url": "https://addyo.substack.com/p/long-running-agents", "publication_date": "2026-04-30", "contribution": "Agents operating across sessions require durable task state, resumability, recovery and verifiable progress.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Persist decisions and current state outside the transient model context; Make retries idempotent and restartable; Verify continuity and completion against external artifacts.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The article examines context-window exhaustion, repeated mistakes and premature \"done\" claims in workflows lasting days. It recommends structured progress artifacts, checkpoints, clean workspaces and independent acceptance tests to support handoffs between agent sessions.", "in_short": "Agents operating across sessions require durable task state, resumability, recovery and verifiable progress.", "main_findings": "Persist decisions and current state outside the transient model context; Make retries idempotent and restartable; Verify continuity and completion against external artifacts.", "bibliography_year": "2026", "section_id": "9.6", "chapter_id": "C09", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W157;W159;W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS013", "stable_bibliography_id": "O013", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "9.5", "manuscript_section_id": "9.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C09\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Long-running agent execution", "K:Agent context and persistent memory", "K:Agent failure and escalation" ], "volumes": [ "V06" ] }, { "id": "O014", "label": "The Agent Stack Bet", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Code quality & security", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C06" ], "printed": false, "summary": "Production agents need identity, authorization, observability and recovery controls beyond an impressive base model.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/the-agent-stack-bet", "authors": "Addy Osmani", "year": "2026-04-18", "attributes": { "label": "The Agent Stack Bet", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Code quality & security", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-agent-stack-bet", "source_url": "https://addyo.substack.com/p/the-agent-stack-bet", "url": "https://addyo.substack.com/p/the-agent-stack-bet", "publication_date": "2026-04-18", "contribution": "Production agents need identity, authorization, observability and recovery controls beyond an impressive base model.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Bind every action to scoped identity and human ownership; Restrict tool permissions and monitor effects; Design recoverable operations and audit trails for runtime failures.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani describes fragile production plumbing: broad service-account permissions, schema drift, opaque state and broken recovery. The remedy is an explicit agent stack with task-scoped access, traceability, policy enforcement and durable error handling.", "in_short": "Production agents need identity, authorization, observability and recovery controls beyond an impressive base model.", "main_findings": "Bind every action to scoped identity and human ownership; Restrict tool permissions and monitor effects; Design recoverable operations and audit trails for runtime failures.", "bibliography_year": "2026", "section_id": "6.8", "chapter_id": "C06", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W053;W060;W152", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS014", "stable_bibliography_id": "O014", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Prompt injection and malicious tools", "K:Dependencies and agent supply chain", "K:Human-agent collaboration" ], "volumes": [] }, { "id": "O015", "label": "Is the IDE dead?", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "Asynchronous agent orchestration may become the primary developer control surface, while editors remain tools for targeted inspection.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/death-of-the-ide", "authors": "Addy Osmani", "year": "2026-03-20", "attributes": { "label": "Is the IDE dead?", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/death-of-the-ide", "source_url": "https://addyo.substack.com/p/death-of-the-ide", "url": "https://addyo.substack.com/p/death-of-the-ide", "publication_date": "2026-03-20", "contribution": "Asynchronous agent orchestration may become the primary developer control surface, while editors remain tools for targeted inspection.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Teach delegation and inspection as separate activities; Compare orchestration tools on observable state and control; Treat claims about the IDE dying as forecasts.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani discusses workbench and control-plane interfaces for coordinating cloud and local coding agents. The claim is about a shift in where engineers plan, inspect and approve work, not proof that code editors will disappear.", "in_short": "Asynchronous agent orchestration may become the primary developer control surface, while editors remain tools for targeted inspection.", "main_findings": "Teach delegation and inspection as separate activities; Compare orchestration tools on observable state and control; Treat claims about the IDE dying as forecasts.", "bibliography_year": "2026", "section_id": "1.3", "chapter_id": "C01", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002;W014;W027", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS015", "stable_bibliography_id": "O015", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Long-running agent execution" ], "volumes": [] }, { "id": "O016", "label": "14 More lessons from 14 years at Google", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "Long-lived engineering teams depend on trust, clear decision-making, effective communication and organizational stewardship.", "limitation": "Personal career retrospective; no original AI-specific field experiment.", "url": "https://addyo.substack.com/p/14-more-lessons-from-14-years-at", "authors": "Addy Osmani", "year": "2026-02-12", "attributes": { "label": "14 More lessons from 14 years at Google", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/14-more-lessons-from-14-years-at", "source_url": "https://addyo.substack.com/p/14-more-lessons-from-14-years-at", "url": "https://addyo.substack.com/p/14-more-lessons-from-14-years-at", "publication_date": "2026-02-12", "contribution": "Long-lived engineering teams depend on trust, clear decision-making, effective communication and organizational stewardship.", "limitation": "Personal career retrospective; no original AI-specific field experiment.", "manual_use": "Make decision rights and handoffs visible; Value team reliability above heroic individual throughput; Use as background principles, not an AI impact study.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "A second retrospective about long-term team effectiveness stresses the structures around projects: communication, ownership, priorities, collaboration and shared decisions. These lessons provide organizational context rather than new research on AI agents.", "in_short": "Long-lived engineering teams depend on trust, clear decision-making, effective communication and organizational stewardship.", "main_findings": "Make decision rights and handoffs visible; Value team reliability above heroic individual throughput; Use as background principles, not an AI impact study.", "bibliography_year": "2026", "section_id": "11.2", "chapter_id": "C11", "relevance": "Adjacent (general engineering)", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W161;W171;W178", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Personal career retrospective; no original AI-specific field experiment. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONTEXTUAL_ONLY", "article_is_ai_specific": "false", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Personal career retrospective; no original AI-specific field experiment. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "editorial_revision_detail": "Remove unnecessary human-agent research implication from a retrospective about teams.; Avoid an unsupported AI comparison inside a retrospective that does not evaluate agents.; Conspicuously prevent non-AI engineering essays from being reported as evidence of AI-specific outcomes.", "prior_record_id": "AOS016", "stable_bibliography_id": "O016", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Ownership and knowledge transfer", "K:Skills formation and teaching", "K:Human cognitive load" ], "volumes": [] }, { "id": "O017", "label": "The 80% Problem in Agentic Coding", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory" ], "actual": [ "C:C10" ], "mapped": [ "C:C10" ], "printed": true, "summary": "When agents implement most code, developers can accumulate comprehension debt even while shipping more changes.", "limitation": "The 80% figure expresses an illustrative workflow shift; it is not a population-level experimental estimate.", "url": "https://addyo.substack.com/p/the-80-problem-in-agentic-coding", "authors": "Addy Osmani", "year": "2026-01-28", "attributes": { "label": "The 80% Problem in Agentic Coding", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-80-problem-in-agentic-coding", "source_url": "https://addyo.substack.com/p/the-80-problem-in-agentic-coding", "url": "https://addyo.substack.com/p/the-80-problem-in-agentic-coding", "publication_date": "2026-01-28", "contribution": "When agents implement most code, developers can accumulate comprehension debt even while shipping more changes.", "limitation": "The 80% figure expresses an illustrative workflow shift; it is not a population-level experimental estimate.", "manual_use": "Require explain-back, behavioral traces and repair exercises; Measure whether an engineer can modify agent-written code later; Treat the 80% phrase as illustration, not a measured universal ratio.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Extending his earlier \"70% problem\", Osmani frames agent-heavy development as an inversion between authorship and understanding. Speed moves the bottleneck toward explaining decisions, tracing behavior and maintaining the code after the generation session ends.", "in_short": "When agents implement most code, developers can accumulate comprehension debt even while shipping more changes.", "main_findings": "Require explain-back, behavioral traces and repair exercises; Measure whether an engineer can modify agent-written code later; Treat the 80% phrase as illustration, not a measured universal ratio.", "bibliography_year": "2026", "section_id": "10.8", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W154;W117;W122", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "The 80% figure expresses an illustrative workflow shift; it is not a population-level experimental estimate.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "The 80% figure expresses an illustrative workflow shift; it is not a population-level experimental estimate.", "editorial_revision_detail": "", "prior_record_id": "AOS017", "stable_bibliography_id": "O017", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "10.2", "manuscript_section_id": "10.8", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C10\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Cognitive offloading and skill debt", "K:System understanding and rationale", "K:Technical debt and complexity" ], "volumes": [ "V06" ] }, { "id": "O018", "label": "How to write a good spec for AI agents", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory" ], "actual": [ "C:C04" ], "mapped": [ "C:C04" ], "printed": true, "summary": "A useful agent specification is compact, testable and iterative rather than a giant prompt resembling an unvalidated requirements dump.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/how-to-write-a-good-spec-for-ai-agents", "authors": "Addy Osmani", "year": "2026-01-19", "attributes": { "label": "How to write a good spec for AI agents", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/how-to-write-a-good-spec-for-ai-agents", "source_url": "https://addyo.substack.com/p/how-to-write-a-good-spec-for-ai-agents", "url": "https://addyo.substack.com/p/how-to-write-a-good-spec-for-ai-agents", "publication_date": "2026-01-19", "contribution": "A useful agent specification is compact, testable and iterative rather than a giant prompt resembling an unvalidated requirements dump.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Specify desired behavior and forbidden regressions separately from implementation detail; Plan before writes, then build in small independently checked increments; Update specification when evidence reveals an ambiguity.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani provides guidance on the degree of detail in requirements, decomposition into small tasks, read-only planning and successive execution. Intent, constraints, expected behavior and testing guidance become checkpoints throughout development.", "in_short": "A useful agent specification is compact, testable and iterative rather than a giant prompt resembling an unvalidated requirements dump.", "main_findings": "Specify desired behavior and forbidden regressions separately from implementation detail; Plan before writes, then build in small independently checked increments; Update specification when evidence reveals an ambiguity.", "bibliography_year": "2026", "section_id": "4.1", "chapter_id": "C04", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W148;W151;W077", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS018", "stable_bibliography_id": "O018", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "4.3", "manuscript_section_id": "4.1", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C04\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Specifications and governance", "K:Agent context and persistent memory", "K:Independent verification and oracles" ], "volumes": [ "V06" ] }, { "id": "O019", "label": "Code Review in the Age of AI", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C08" ], "printed": false, "summary": "The rise in generated changes makes code review an explicit burden-of-proof and accountability process.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/code-review-in-the-age-of-ai", "authors": "Addy Osmani", "year": "2026-01-05", "attributes": { "label": "Code Review in the Age of AI", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/code-review-in-the-age-of-ai", "source_url": "https://addyo.substack.com/p/code-review-in-the-age-of-ai", "url": "https://addyo.substack.com/p/code-review-in-the-age-of-ai", "publication_date": "2026-01-05", "contribution": "The rise in generated changes makes code review an explicit burden-of-proof and accountability process.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Attach objective verification artifacts to pull requests; Review intent, integration risk and sensitive paths; Never treat a passing self-generated test as independent approval.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay argues that tests, reproducible manual checks and a stated risk assessment should accompany every proposal. Teams should preserve contextual knowledge and ownership rather than confusing a model-generated diff with evidence that the resulting change is safe.", "in_short": "The rise in generated changes makes code review an explicit burden-of-proof and accountability process.", "main_findings": "Attach objective verification artifacts to pull requests; Review intent, integration risk and sensitive paths; Never treat a passing self-generated test as independent approval.", "bibliography_year": "2026", "section_id": "8.9", "chapter_id": "C08", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W065;W069;W093", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2026", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "editorial_revision_detail": "Secondhand empirical or vendor statistics must not be mistaken for independently replicated studies.", "prior_record_id": "AOS019", "stable_bibliography_id": "O019", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Review capacity and maintainers", "K:Independent verification and oracles", "K:Ownership and knowledge transfer" ], "volumes": [] }, { "id": "O020", "label": "How Good Is AI at Coding React (Really)?", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "React-related AI success is highly task-dependent, with integration and usability much harder to establish than component scaffolding.", "limitation": "Numerical benchmark claims are a synthesis of separate benchmarks with different tasks, versions and measurement methods; not directly comparable.", "url": "https://addyo.substack.com/p/how-good-is-ai-at-coding-react-really", "authors": "Addy Osmani", "year": "2025-12-29", "attributes": { "label": "How Good Is AI at Coding React (Really)?", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/how-good-is-ai-at-coding-react-really", "source_url": "https://addyo.substack.com/p/how-good-is-ai-at-coding-react-really", "url": "https://addyo.substack.com/p/how-good-is-ai-at-coding-react-really", "publication_date": "2025-12-29", "contribution": "React-related AI success is highly task-dependent, with integration and usability much harder to establish than component scaffolding.", "limitation": "Numerical benchmark claims are a synthesis of separate benchmarks with different tasks, versions and measurement methods; not directly comparable.", "manual_use": "Benchmark scores have task- and version-specific denominators; Use browser tests and accessibility review for end-to-end UI changes; Treat reported percentages as third-party snapshot results, not the author’s experimental measurement.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Using third-party web and coding benchmarks, Osmani contrasts relatively tractable isolated UI tasks with weaker multi-step integration. He recommends context-rich specifications, UI standards, accessibility checks and browser-level evaluation instead of equating code generation with product quality.", "in_short": "React-related AI success is highly task-dependent, with integration and usability much harder to establish than component scaffolding.", "main_findings": "Benchmark scores have task- and version-specific denominators; Use browser tests and accessibility review for end-to-end UI changes; Treat reported percentages as third-party snapshot results, not the author’s experimental measurement.", "bibliography_year": "2025", "section_id": "3.4", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W191;W193;W111", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Numerical benchmark claims are a synthesis of separate benchmarks with different tasks, versions and measurement methods; not directly comparable. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Numerical benchmark claims are a synthesis of separate benchmarks with different tasks, versions and measurement methods; not directly comparable. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "editorial_revision_detail": "Secondhand empirical or vendor statistics must not be mistaken for independently replicated studies.", "prior_record_id": "AOS020", "stable_bibliography_id": "O020", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Benchmarks and construct validity", "K:Agent failure and escalation", "K:Architecture and modularity" ], "volumes": [] }, { "id": "O021", "label": "My LLM coding workflow going into 2026", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04" ], "printed": false, "summary": "Planning, scoped delegation and continuous checking outperform an unmanaged prompt-to-merge workflow as an engineering practice.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/my-llm-coding-workflow-going-into", "authors": "Addy Osmani", "year": "2025-12-18", "attributes": { "label": "My LLM coding workflow going into 2026", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/my-llm-coding-workflow-going-into", "source_url": "https://addyo.substack.com/p/my-llm-coding-workflow-going-into", "url": "https://addyo.substack.com/p/my-llm-coding-workflow-going-into", "publication_date": "2025-12-18", "contribution": "Planning, scoped delegation and continuous checking outperform an unmanaged prompt-to-merge workflow as an engineering practice.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Start from a task brief and acceptance criteria; Review after each small increment; Distinguish personal workflow preference from measured performance gains.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani describes his own evolving approach to working with coding assistants: articulate a plan, provide repository context, implement bounded steps, execute tests and remain accountable for reviews and integration.", "in_short": "Planning, scoped delegation and continuous checking outperform an unmanaged prompt-to-merge workflow as an engineering practice.", "main_findings": "Start from a task brief and acceptance criteria; Review after each small increment; Distinguish personal workflow preference from measured performance gains.", "bibliography_year": "2025", "section_id": "4.6", "chapter_id": "C04", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W151;W095;W083", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS021", "stable_bibliography_id": "O021", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Specifications and governance", "K:Developer workflow and adoption", "K:Independent verification and oracles" ], "volumes": [] }, { "id": "O022", "label": "21 Lessons from 14 Years at Google", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "Durable engineering effectiveness depends on solving user problems, judgment and collaboration beyond producing code.", "limitation": "Career reflection; not original AI outcome research.", "url": "https://addyo.substack.com/p/21-lessons-from-14-years-at-google", "authors": "Addy Osmani", "year": "2025-12-04", "attributes": { "label": "21 Lessons from 14 Years at Google", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/21-lessons-from-14-years-at-google", "source_url": "https://addyo.substack.com/p/21-lessons-from-14-years-at-google", "url": "https://addyo.substack.com/p/21-lessons-from-14-years-at-google", "publication_date": "2025-12-04", "contribution": "Durable engineering effectiveness depends on solving user problems, judgment and collaboration beyond producing code.", "limitation": "Career reflection; not original AI outcome research.", "manual_use": "Optimize for user outcomes instead of code volume; Preserve communication, shared understanding and professional judgment; Keep this as general engineering context.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani shares lessons from his work at Google about user empathy, simplifying problems, team alignment and career growth. The lessons supply an organizational foundation for AI-assisted engineering without presenting direct evidence about coding agents.", "in_short": "Durable engineering effectiveness depends on solving user problems, judgment and collaboration beyond producing code.", "main_findings": "Optimize for user outcomes instead of code volume; Preserve communication, shared understanding and professional judgment; Keep this as general engineering context.", "bibliography_year": "2025", "section_id": "11.2", "chapter_id": "C11", "relevance": "Adjacent (general engineering)", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W161;W175;W184", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Career reflection; not original AI outcome research. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONTEXTUAL_ONLY", "article_is_ai_specific": "false", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Career reflection; not original AI outcome research. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "editorial_revision_detail": "General team/career essay cannot itself establish human-agent collaboration effects.; Conspicuously prevent non-AI engineering essays from being reported as evidence of AI-specific outcomes.", "prior_record_id": "AOS022", "stable_bibliography_id": "O022", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Ownership and knowledge transfer", "K:System understanding and rationale", "K:Skills formation and teaching" ], "volumes": [] }, { "id": "O023", "label": "Treat AI-Generated code as a draft", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:System knowledge & memory" ], "actual": [ "C:C08" ], "mapped": [ "C:C08" ], "printed": true, "summary": "AI-generated code should be treated as an untrusted draft and not merged without responsible understanding and verification.", "limitation": "Prescriptive practice subsequently qualified by the same author in September 2026.", "url": "https://addyo.substack.com/p/treat-ai-generated-code-as-a-draft", "authors": "Addy Osmani", "year": "2025-11-25", "attributes": { "label": "Treat AI-Generated code as a draft", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/treat-ai-generated-code-as-a-draft", "source_url": "https://addyo.substack.com/p/treat-ai-generated-code-as-a-draft", "url": "https://addyo.substack.com/p/treat-ai-generated-code-as-a-draft", "publication_date": "2025-11-25", "contribution": "AI-generated code should be treated as an untrusted draft and not merged without responsible understanding and verification.", "limitation": "Prescriptive practice subsequently qualified by the same author in September 2026.", "manual_use": "Verify AI output against requirements before acceptance; Record decision ownership; Document the change in the author’s proposed review policy over time.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "This earlier essay stresses reading AI code closely so that humans connect implementation back to intent, testing and maintenance. Its strong line-by-line reading prescription is explicitly revised in the September 2026 \"The Code Nobody Reads\" essay, which advocates risk-tiered independent checking.", "in_short": "AI-generated code should be treated as an untrusted draft and not merged without responsible understanding and verification.", "main_findings": "Verify AI output against requirements before acceptance; Record decision ownership; Document the change in the author’s proposed review policy over time.", "bibliography_year": "2025", "section_id": "8.8", "chapter_id": "C08", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W067;W069;W154", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Prescriptive practice subsequently qualified by the same author in September 2026.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Prescriptive practice subsequently qualified by the same author in September 2026.", "editorial_revision_detail": "", "prior_record_id": "AOS023", "stable_bibliography_id": "O023", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "8.7", "manuscript_section_id": "8.8", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C08\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Independent verification and oracles", "K:System understanding and rationale", "K:Ownership and knowledge transfer" ], "volumes": [ "V06" ] }, { "id": "O024", "label": "Critical Thinking during the age of AI", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C10" ], "printed": false, "summary": "Critical questions about purpose, assumptions, context and evidence remain necessary when AI supplies plausible solutions.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/critical-thinking-during-the-age", "authors": "Addy Osmani", "year": "2025-11-21", "attributes": { "label": "Critical Thinking during the age of AI", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/critical-thinking-during-the-age", "source_url": "https://addyo.substack.com/p/critical-thinking-during-the-age", "url": "https://addyo.substack.com/p/critical-thinking-during-the-age", "publication_date": "2025-11-21", "contribution": "Critical questions about purpose, assumptions, context and evidence remain necessary when AI supplies plausible solutions.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Ask what result is needed before asking for code; Require evidence for assumptions and fixes; Train developers to challenge generated explanations.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani organizes an engineering reasoning checklist around who, what, where, when, why and how. The article emphasizes problem definition, root-cause analysis and skepticism toward well-phrased AI output.", "in_short": "Critical questions about purpose, assumptions, context and evidence remain necessary when AI supplies plausible solutions.", "main_findings": "Ask what result is needed before asking for code; Require evidence for assumptions and fixes; Train developers to challenge generated explanations.", "bibliography_year": "2025", "section_id": "10.5", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W117;W118", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "A biometric AI-coding study is distant from an essay primarily recommending analytical questioning and critical reasoning.", "prior_record_id": "AOS024", "stable_bibliography_id": "O024", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Human cognitive load", "K:Skills formation and teaching", "K:Specifications and governance" ], "volumes": [] }, { "id": "O025", "label": "Conductors to Orchestrators: The Future of Agentic Coding", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Jobs & roles" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "Developers are moving from guiding one agent toward coordinating multiple asynchronously executing agents.", "limitation": "Role forecasts and workflow examples, not validated hiring or productivity outcomes.", "url": "https://addyo.substack.com/p/conductors-to-orchestrators-the-future", "authors": "Addy Osmani", "year": "2025-11-01", "attributes": { "label": "Conductors to Orchestrators: The Future of Agentic Coding", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/conductors-to-orchestrators-the-future", "source_url": "https://addyo.substack.com/p/conductors-to-orchestrators-the-future", "url": "https://addyo.substack.com/p/conductors-to-orchestrators-the-future", "publication_date": "2025-11-01", "contribution": "Developers are moving from guiding one agent toward coordinating multiple asynchronously executing agents.", "limitation": "Role forecasts and workflow examples, not validated hiring or productivity outcomes.", "manual_use": "Isolate concurrent branches and agent responsibilities; Assign merge and conflict-resolution ownership; Treat role changes as a scenario to evaluate, not an established labor-market effect.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay distinguishes a single-agent \"conductor\" role from an \"orchestrator\" that allocates work among parallel agents with isolated workspaces. It forecasts changes in the engineer’s day-to-day role while acknowledging tool maturity and coordination costs.", "in_short": "Developers are moving from guiding one agent toward coordinating multiple asynchronously executing agents.", "main_findings": "Isolate concurrent branches and agent responsibilities; Assign merge and conflict-resolution ownership; Treat role changes as a scenario to evaluate, not an established labor-market effect.", "bibliography_year": "2025", "section_id": "11.2", "chapter_id": "C11", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002;W027;W152", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Role forecasts and workflow examples, not validated hiring or productivity outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Role forecasts and workflow examples, not validated hiring or productivity outcomes.", "editorial_revision_detail": "", "prior_record_id": "AOS025", "stable_bibliography_id": "O025", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Human-agent collaboration", "K:Long-running agent execution", "K:Developer workflow and adoption" ], "volumes": [] }, { "id": "O026", "label": "Gemini CLI Tips & Tricks", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "Command-line agent effectiveness depends on tool configuration, project context and deliberate execution controls.", "limitation": "Vendor-tool feature guide; installation flags and quotas may have changed since publication.", "url": "https://addyo.substack.com/p/gemini-cli-tips-and-tricks", "authors": "Addy Osmani", "year": "2025-10-21", "attributes": { "label": "Gemini CLI Tips & Tricks", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/gemini-cli-tips-and-tricks", "source_url": "https://addyo.substack.com/p/gemini-cli-tips-and-tricks", "url": "https://addyo.substack.com/p/gemini-cli-tips-and-tricks", "publication_date": "2025-10-21", "contribution": "Command-line agent effectiveness depends on tool configuration, project context and deliberate execution controls.", "limitation": "Vendor-tool feature guide; installation flags and quotas may have changed since publication.", "manual_use": "Teach observable tool permissions and commands; Keep version-specific instructions separate from timeless engineering principles; Do not infer model superiority from a tips guide.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "A hands-on guide to Gemini CLI installation, task prompting, tool use and operational shortcuts. Its practical examples illustrate the capabilities and risks of agent access to shells, repositories and developer data.", "in_short": "Command-line agent effectiveness depends on tool configuration, project context and deliberate execution controls.", "main_findings": "Teach observable tool permissions and commands; Keep version-specific instructions separate from timeless engineering principles; Do not infer model superiority from a tips guide.", "bibliography_year": "2025", "section_id": "3.2", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Vendor-tool feature guide; installation flags and quotas may have changed since publication.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Vendor-tool feature guide; installation flags and quotas may have changed since publication.", "editorial_revision_detail": "Early-2026 Microsoft adoption and a specialized virtual-memory design do not verify Gemini CLI’s 2025 tool features; retain only a broad agent-failure comparator.", "prior_record_id": "AOS026", "stable_bibliography_id": "O026", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Agent context and persistent memory", "K:Developer workflow and adoption", "K:Agent failure and escalation" ], "volumes": [] }, { "id": "O027", "label": "How modern browsers work", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure" ], "actual": [], "mapped": [ "C:C05" ], "printed": false, "summary": "Modern browsers coordinate networking, parsing, JavaScript execution, rendering, process isolation and security boundaries.", "limitation": "General browser engineering explainer; no claims of AI development outcomes.", "url": "https://addyo.substack.com/p/how-modern-browsers-work", "authors": "Addy Osmani", "year": "2025-09-13", "attributes": { "label": "How modern browsers work", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/how-modern-browsers-work", "source_url": "https://addyo.substack.com/p/how-modern-browsers-work", "url": "https://addyo.substack.com/p/how-modern-browsers-work", "publication_date": "2025-09-13", "contribution": "Modern browsers coordinate networking, parsing, JavaScript execution, rendering, process isolation and security boundaries.", "limitation": "General browser engineering explainer; no claims of AI development outcomes.", "manual_use": "Use browser internals as a brownfield-comprehension teaching example; Distinguish runtime behavior from source-file appearance; Exclude it from AI-specific evidence claims.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "A broad walkthrough of modern browser internals, including navigation, parsing, JavaScript, GPU rendering, multiprocessing and sandboxing. It is not about AI-assisted engineering; its connection to the graph is contextual architecture literacy.", "in_short": "Modern browsers coordinate networking, parsing, JavaScript execution, rendering, process isolation and security boundaries.", "main_findings": "Use browser internals as a brownfield-comprehension teaching example; Distinguish runtime behavior from source-file appearance; Exclude it from AI-specific evidence claims.", "bibliography_year": "2025", "section_id": "5.5", "chapter_id": "C05", "relevance": "Adjacent (non-AI technical)", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "General browser engineering explainer; no claims of AI development outcomes. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONTEXTUAL_ONLY", "article_is_ai_specific": "false", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "General browser engineering explainer; no claims of AI development outcomes. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "editorial_revision_detail": "The source is a browser-internals overview, not a study of AI-created code quality or security.; No AI research cross-link is sufficiently specific to an article about browser internals; keep for general architectural context only.; The source is specifically an explanation of browser internals, with no discussion of AI-generated modifications.; Conspicuously prevent non-AI engineering essays from being reported as evidence of AI-specific outcomes.", "prior_record_id": "AOS027", "stable_bibliography_id": "O027", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Architecture and modularity", "K:Vulnerability and secure code", "K:System understanding and rationale" ], "volumes": [] }, { "id": "O028", "label": "Vibe coding is not the same as AI-Assisted engineering.", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Code quality & security" ], "actual": [ "C:C04" ], "mapped": [ "C:C04" ], "printed": true, "summary": "Fast conversational prototyping and accountable AI-assisted engineering are distinct development regimes.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/vibe-coding-is-not-the-same-as-ai", "authors": "Addy Osmani", "year": "2025-08-30", "attributes": { "label": "Vibe coding is not the same as AI-Assisted engineering.", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/vibe-coding-is-not-the-same-as-ai", "source_url": "https://addyo.substack.com/p/vibe-coding-is-not-the-same-as-ai", "url": "https://addyo.substack.com/p/vibe-coding-is-not-the-same-as-ai", "publication_date": "2025-08-30", "contribution": "Fast conversational prototyping and accountable AI-assisted engineering are distinct development regimes.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Choose rigor appropriate to the target environment; Require explicit acceptance contracts for production work; Teach the transition from prototype to maintainable system.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani distinguishes exploratory prompting from production practices with design documents, test-driven changes, review and ownership. He argues that treating either workflow as universally appropriate obscures the requirements of maintenance-intensive software.", "in_short": "Fast conversational prototyping and accountable AI-assisted engineering are distinct development regimes.", "main_findings": "Choose rigor appropriate to the target environment; Require explicit acceptance contracts for production work; Teach the transition from prototype to maintainable system.", "bibliography_year": "2025", "section_id": "4.1", "chapter_id": "C04", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W045;W050;W151", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS028", "stable_bibliography_id": "O028", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "4.1", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C04\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Technical debt and complexity" ], "volumes": [ "V06" ] }, { "id": "O029", "label": "The reality of AI-Assisted software engineering productivity", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Productivity", "T:Economics" ], "actual": [ "C:C02" ], "mapped": [ "C:C02" ], "printed": true, "summary": "Perceived personal acceleration with coding AI does not necessarily translate into net team or release-level productivity.", "limitation": "The percentages reported by Osmani come from diverse third-party surveys and experiments; no single pooled causal estimate is supplied.", "url": "https://addyo.substack.com/p/the-reality-of-ai-assisted-software", "authors": "Addy Osmani", "year": "2025-08-16", "attributes": { "label": "The reality of AI-Assisted software engineering productivity", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Productivity", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-reality-of-ai-assisted-software", "source_url": "https://addyo.substack.com/p/the-reality-of-ai-assisted-software", "url": "https://addyo.substack.com/p/the-reality-of-ai-assisted-software", "publication_date": "2025-08-16", "contribution": "Perceived personal acceleration with coding AI does not necessarily translate into net team or release-level productivity.", "limitation": "The percentages reported by Osmani come from diverse third-party surveys and experiments; no single pooled causal estimate is supplied.", "manual_use": "Use task-adjusted and quality-adjusted metrics; Report study context before citing a percentage; Separate survey sentiment, experiments and shipped-value outcomes.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The article synthesizes surveys and selected field studies showing variation by task, developer and codebase. It cautions against treating high adoption, code generation or self-reported speed as direct evidence of deployment throughput or organizational value.", "in_short": "Perceived personal acceleration with coding AI does not necessarily translate into net team or release-level productivity.", "main_findings": "Use task-adjusted and quality-adjusted metrics; Report study context before citing a percentage; Separate survey sentiment, experiments and shipped-value outcomes.", "bibliography_year": "2025", "section_id": "2.5", "chapter_id": "C02", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W003;W023;W010", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "The percentages reported by Osmani come from diverse third-party surveys and experiments; no single pooled causal estimate is supplied. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "The percentages reported by Osmani come from diverse third-party surveys and experiments; no single pooled causal estimate is supplied. Any externally quoted benchmark, productivity or vendor telemetry requires independent verification of original method and population.", "editorial_revision_detail": "Secondhand empirical or vendor statistics must not be mistaken for independently replicated studies.", "prior_record_id": "AOS029", "stable_bibliography_id": "O029", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "2.3", "manuscript_section_id": "2.5", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C02\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Controlled productivity evidence", "K:Output versus shipped value", "K:Evaluation methodology and baselines" ], "volumes": [ "V06" ] }, { "id": "O030", "label": "Coding for the Future Agentic World", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "CLI agents and background task runners extend software generation into tool-using, multi-step workflows, but require more oversight.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/coding-for-the-future-agentic-world", "authors": "Addy Osmani", "year": "2025-07-29", "attributes": { "label": "Coding for the Future Agentic World", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/coding-for-the-future-agentic-world", "source_url": "https://addyo.substack.com/p/coding-for-the-future-agentic-world", "url": "https://addyo.substack.com/p/coding-for-the-future-agentic-world", "publication_date": "2025-07-29", "contribution": "CLI agents and background task runners extend software generation into tool-using, multi-step workflows, but require more oversight.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Teach capabilities with bounded task examples; Evaluate tool use and integration separately from model fluency; Retain explicit sandbox and final release authority.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani surveys emerging agentic tools and contrasts autocomplete, interactive assistants and agents that edit files, run commands, debug and propose pull requests. He argues that more capable execution increases the value of planning, context and final approval.", "in_short": "CLI agents and background task runners extend software generation into tool-using, multi-step workflows, but require more oversight.", "main_findings": "Teach capabilities with bounded task examples; Evaluate tool use and integration separately from model fluency; Retain explicit sandbox and final release authority.", "bibliography_year": "2025", "section_id": "3.1", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002;W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "A 2026 agent-adoption study postdates this essay and is not evidence for its original predictions. Keep process/limitations comparators.", "prior_record_id": "AOS030", "stable_bibliography_id": "O030", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Long-running agent execution", "K:Agent failure and escalation" ], "volumes": [] }, { "id": "O031", "label": "Context Engineering: Bringing Engineering Discipline to Prompts", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:System knowledge & memory", "T:Engineering structure" ], "actual": [ "C:C09" ], "mapped": [ "C:C09" ], "printed": true, "summary": "Reliable prompts are assembled from selected system context, tools and constraints rather than crafted by wording alone.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/context-engineering-bringing-engineering", "authors": "Addy Osmani", "year": "2025-07-13", "attributes": { "label": "Context Engineering: Bringing Engineering Discipline to Prompts", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "System knowledge & memory", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/context-engineering-bringing-engineering", "source_url": "https://addyo.substack.com/p/context-engineering-bringing-engineering", "url": "https://addyo.substack.com/p/context-engineering-bringing-engineering", "publication_date": "2025-07-13", "contribution": "Reliable prompts are assembled from selected system context, tools and constraints rather than crafted by wording alone.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Treat context as a versioned engineering input; Prefer relevant artifacts over exhaustive dumps; Verify provenance and freshness of retrieved context.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay develops context engineering as information architecture: relevant files, architectural records, error logs, schemas, examples and acceptance constraints. It emphasizes curating high-signal context so a coding agent can act within repository realities.", "in_short": "Reliable prompts are assembled from selected system context, tools and constraints rather than crafted by wording alone.", "main_findings": "Treat context as a versioned engineering input; Prefer relevant artifacts over exhaustive dumps; Verify provenance and freshness of retrieved context.", "bibliography_year": "2025", "section_id": "9.6", "chapter_id": "C09", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W155;W159;W184", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS031", "stable_bibliography_id": "O031", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "original_editorial_section_id": "9.4", "manuscript_section_id": "9.6", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C09\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Specifications and governance" ], "volumes": [ "V06" ] }, { "id": "O032", "label": "The AI-Native Software Engineer", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C10" ], "printed": false, "summary": "An AI-native engineer integrates assistance into ordinary work while retaining domain expertise and continuous learning.", "limitation": "Prospective productivity multipliers are hypothetical and should not be cited as empirical measurements.", "url": "https://addyo.substack.com/p/the-ai-native-software-engineer", "authors": "Addy Osmani", "year": "2025-07-01", "attributes": { "label": "The AI-Native Software Engineer", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-ai-native-software-engineer", "source_url": "https://addyo.substack.com/p/the-ai-native-software-engineer", "url": "https://addyo.substack.com/p/the-ai-native-software-engineer", "publication_date": "2025-07-01", "contribution": "An AI-native engineer integrates assistance into ordinary work while retaining domain expertise and continuous learning.", "limitation": "Prospective productivity multipliers are hypothetical and should not be cited as empirical measurements.", "manual_use": "Teach effective delegation alongside manual competence; Create deliberate skill-updating routines; Do not present advertised 2x–10x multipliers as validated general results.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani presents a practical skills playbook for pairing with coding assistants, delegating repetitive work and learning better tool use. The article is intentionally aspirational about possible productivity multiples rather than a controlled performance study.", "in_short": "An AI-native engineer integrates assistance into ordinary work while retaining domain expertise and continuous learning.", "main_findings": "Teach effective delegation alongside manual competence; Create deliberate skill-updating routines; Do not present advertised 2x–10x multipliers as validated general results.", "bibliography_year": "2025", "section_id": "10.9", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W116;W118;W122", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Prospective productivity multipliers are hypothetical and should not be cited as empirical measurements.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Prospective productivity multipliers are hypothetical and should not be cited as empirical measurements.", "editorial_revision_detail": "", "prior_record_id": "AOS032", "stable_bibliography_id": "O032", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Skills formation and teaching", "K:Human-agent collaboration" ], "volumes": [] }, { "id": "O033", "label": "The \"Trust, But Verify\" Pattern For AI-Assisted Engineering", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C07" ], "printed": false, "summary": "Verification becomes the bottleneck when generation is cheap; every AI-produced change needs a justified check.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/the-trust-but-verify-pattern-for", "authors": "Addy Osmani", "year": "2025-06-09", "attributes": { "label": "The \"Trust, But Verify\" Pattern For AI-Assisted Engineering", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-trust-but-verify-pattern-for", "source_url": "https://addyo.substack.com/p/the-trust-but-verify-pattern-for", "url": "https://addyo.substack.com/p/the-trust-but-verify-pattern-for", "publication_date": "2025-06-09", "contribution": "Verification becomes the bottleneck when generation is cheap; every AI-produced change needs a justified check.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Check against independent requirements and regression oracles; Record the outcome of manual and automated review; Teach why plausible code is not proof of correctness.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani frames the assistant as a capable but fallible collaborator whose output must be tested and reviewed before adoption. The post translates automation-bias concerns into reviewer practices, guardrails and accountable sign-off.", "in_short": "Verification becomes the bottleneck when generation is cheap; every AI-produced change needs a justified check.", "main_findings": "Check against independent requirements and regression oracles; Record the outcome of manual and automated review; Teach why plausible code is not proof of correctness.", "bibliography_year": "2025", "section_id": "7.2", "chapter_id": "C07", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W067;W077;W093", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS033", "stable_bibliography_id": "O033", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Independent verification and oracles", "K:Automation bias and trust", "K:Review capacity and maintainers" ], "volumes": [] }, { "id": "O034", "label": "The Prompt Engineering Playbook for Programmers", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04" ], "printed": false, "summary": "Effective developer prompts state a bounded problem, relevant context, constraints and the expected verification method.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/the-prompt-engineering-playbook-for", "authors": "Addy Osmani", "year": "2025-05-27", "attributes": { "label": "The Prompt Engineering Playbook for Programmers", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-prompt-engineering-playbook-for", "source_url": "https://addyo.substack.com/p/the-prompt-engineering-playbook-for", "url": "https://addyo.substack.com/p/the-prompt-engineering-playbook-for", "publication_date": "2025-05-27", "contribution": "Effective developer prompts state a bounded problem, relevant context, constraints and the expected verification method.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Provide concrete files and constraints with each request; Make successful output observable; Iterate with feedback instead of claiming prompt wording ensures reliability.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Through contrasting examples, Osmani demonstrates how explicit role, environment, scope, format and feedback loops can improve coding-assistant usefulness. The book can reuse the patterns as task-envelope teaching devices rather than treat prompt form as a substitute for formal specifications.", "in_short": "Effective developer prompts state a bounded problem, relevant context, constraints and the expected verification method.", "main_findings": "Provide concrete files and constraints with each request; Make successful output observable; Iterate with feedback instead of claiming prompt wording ensures reliability.", "bibliography_year": "2025", "section_id": "4.3", "chapter_id": "C04", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W151;W077;W148", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "The prompt playbook teaches expected behavior and context of task specifications rather than only versioning an intention ledger.; Comprehension of unfamiliar code (W155) is too remote from creating prompts with specification and verifier constraints.", "prior_record_id": "AOS034", "stable_bibliography_id": "O034", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Specifications and governance", "K:Agent context and persistent memory", "K:Developer workflow and adoption" ], "volumes": [] }, { "id": "O035", "label": "AI Won't Kill Junior Devs - But Your Hiring Strategy Might", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Jobs & roles", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C12" ], "printed": false, "summary": "Removing junior engineering positions can damage apprenticeship and the future pool of experienced maintainers.", "limitation": "A prescriptive career and hiring essay; no original causal analysis of employment trends.", "url": "https://addyo.substack.com/p/ai-wont-kill-junior-devs-but-your", "authors": "Addy Osmani", "year": "2025-05-10", "attributes": { "label": "AI Won't Kill Junior Devs - But Your Hiring Strategy Might", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Jobs & roles", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/ai-wont-kill-junior-devs-but-your", "source_url": "https://addyo.substack.com/p/ai-wont-kill-junior-devs-but-your", "url": "https://addyo.substack.com/p/ai-wont-kill-junior-devs-but-your", "publication_date": "2025-05-10", "contribution": "Removing junior engineering positions can damage apprenticeship and the future pool of experienced maintainers.", "limitation": "A prescriptive career and hiring essay; no original causal analysis of employment trends.", "manual_use": "Protect supervised learning opportunities for junior developers; Evaluate apprenticeship capacity alongside coding throughput; Compare hiring claims with primary labor-market studies.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani argues that early-career roles must evolve toward debugging, system comprehension, collaboration and verification as assistants automate routine output. This is a workforce-development argument, not a causal demonstration that AI has or has not changed entry-level hiring.", "in_short": "Removing junior engineering positions can damage apprenticeship and the future pool of experienced maintainers.", "main_findings": "Protect supervised learning opportunities for junior developers; Evaluate apprenticeship capacity alongside coding throughput; Compare hiring claims with primary labor-market studies.", "bibliography_year": "2025", "section_id": "12.7", "chapter_id": "C12", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W112;W125;W131", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "A prescriptive career and hiring essay; no original causal analysis of employment trends.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "A prescriptive career and hiring essay; no original causal analysis of employment trends.", "editorial_revision_detail": "The argument is explicitly about the junior labor pipeline and apprenticeship economics; C12.7 is a closer primary home than team staffing mechanics.; The argument is explicitly about the junior labor pipeline and apprenticeship economics; C12.7 is a closer primary home than team staffing mechanics.", "prior_record_id": "AOS035", "stable_bibliography_id": "O035", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Employment and early-career roles", "K:Skills formation and teaching", "K:Onboarding and developer fluency" ], "volumes": [] }, { "id": "O036", "label": "Avoiding Skill Atrophy in the Age of AI", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C10" ], "printed": false, "summary": "Using AI to avoid every difficult step risks eroding the reasoning skills needed to supervise future AI output.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/avoiding-skill-atrophy-in-the-age", "authors": "Addy Osmani", "year": "2025-04-25", "attributes": { "label": "Avoiding Skill Atrophy in the Age of AI", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/avoiding-skill-atrophy-in-the-age", "source_url": "https://addyo.substack.com/p/avoiding-skill-atrophy-in-the-age", "url": "https://addyo.substack.com/p/avoiding-skill-atrophy-in-the-age", "publication_date": "2025-04-25", "contribution": "Using AI to avoid every difficult step risks eroding the reasoning skills needed to supervise future AI output.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Maintain opportunities for unaided practice; Use delayed comprehension and transfer tests; Keep the analogy separate from evidence of measured developer skill decay.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani draws an analogy with other forms of cognitive offloading and recommends intentional debugging, independent implementation, explanation and review. The essay identifies a plausible risk, while the size and persistence of skill loss require empirical evaluation.", "in_short": "Using AI to avoid every difficult step risks eroding the reasoning skills needed to supervise future AI output.", "main_findings": "Maintain opportunities for unaided practice; Use delayed comprehension and transfer tests; Keep the analogy separate from evidence of measured developer skill decay.", "bibliography_year": "2025", "section_id": "10.4", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W117;W118;W121", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS036", "stable_bibliography_id": "O036", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Cognitive offloading and skill debt", "K:Skills formation and teaching", "K:Onboarding and developer fluency" ], "volumes": [] }, { "id": "O037", "label": "Vibe Coding is not an excuse for low-quality work", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Code quality & security", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C05" ], "printed": false, "summary": "Rapid AI prototyping does not exempt production changes from ordinary software quality obligations.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/vibe-coding-is-not-an-excuse-for", "authors": "Addy Osmani", "year": "2025-04-18", "attributes": { "label": "Vibe Coding is not an excuse for low-quality work", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Code quality & security", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/vibe-coding-is-not-an-excuse-for", "source_url": "https://addyo.substack.com/p/vibe-coding-is-not-an-excuse-for", "url": "https://addyo.substack.com/p/vibe-coding-is-not-an-excuse-for", "publication_date": "2025-04-18", "contribution": "Rapid AI prototyping does not exempt production changes from ordinary software quality obligations.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Evaluate generated artifacts against system standards; Require a maintenance plan for promoted prototypes; Measure longer-run defects and complexity, not demo velocity.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani warns that easy generation can conceal bugs, weak design and unmaintainable dependencies. He urges standard review, architectural judgment, tests and security checks even when non-specialists can produce working prototypes.", "in_short": "Rapid AI prototyping does not exempt production changes from ordinary software quality obligations.", "main_findings": "Evaluate generated artifacts against system standards; Require a maintenance plan for promoted prototypes; Measure longer-run defects and complexity, not demo velocity.", "bibliography_year": "2025", "section_id": "5.8", "chapter_id": "C05", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W039;W050;W059", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS037", "stable_bibliography_id": "O037", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Technical debt and complexity", "K:Independent verification and oracles", "K:Architecture and modularity" ], "volumes": [] }, { "id": "O038", "label": "What are AI Agents? why do they matter?", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "An agent differs from one-shot model completion by using tools and feedback to pursue a goal over multiple steps.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/what-are-ai-agents-why-do-they-matter", "authors": "Addy Osmani", "year": "2025-04-11", "attributes": { "label": "What are AI Agents? why do they matter?", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/what-are-ai-agents-why-do-they-matter", "source_url": "https://addyo.substack.com/p/what-are-ai-agents-why-do-they-matter", "url": "https://addyo.substack.com/p/what-are-ai-agents-why-do-they-matter", "publication_date": "2025-04-11", "contribution": "An agent differs from one-shot model completion by using tools and feedback to pursue a goal over multiple steps.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Model agents as bounded goal-directed processes; Separate tool authority from language-model output; Require observable success and failure states.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The post offers an introductory description of agent perception, planning, tool use, action and iteration. It is useful for defining system boundaries and distinguishing model capability from the reliability of a surrounding task workflow.", "in_short": "An agent differs from one-shot model completion by using tools and feedback to pursue a goal over multiple steps.", "main_findings": "Model agents as bounded goal-directed processes; Separate tool authority from language-model output; Require observable success and failure states.", "bibliography_year": "2025", "section_id": "3.1", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W191;W002;W152", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "AIDev (W192) supplies a PR dataset, rather than evidence about the conceptual structure of AI agents; compare workflow studies.", "prior_record_id": "AOS038", "stable_bibliography_id": "O038", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Human-agent collaboration" ], "volumes": [] }, { "id": "O039", "label": "Vibe Coding: Revolution or Reckless Abandon?", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Code quality & security", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C05" ], "printed": false, "summary": "Vibe coding can democratize prototypes while amplifying unnoticed security and maintainability risks.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/vibe-coding-revolution-or-reckless", "authors": "Addy Osmani", "year": "2025-04-03", "attributes": { "label": "Vibe Coding: Revolution or Reckless Abandon?", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Code quality & security", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/vibe-coding-revolution-or-reckless", "source_url": "https://addyo.substack.com/p/vibe-coding-revolution-or-reckless", "url": "https://addyo.substack.com/p/vibe-coding-revolution-or-reckless", "publication_date": "2025-04-03", "contribution": "Vibe coding can democratize prototypes while amplifying unnoticed security and maintainability risks.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Separate creative prototyping from production acceptance; Add security and regression checks early; Train users to recognize the limits of superficial functional demos.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani discusses enthusiastic prompting with minimal code inspection and contrasts its accessibility with risks from incorrect assumptions, latent defects and insufficient operator understanding. The essay calls for professional checks as project consequence increases.", "in_short": "Vibe coding can democratize prototypes while amplifying unnoticed security and maintainability risks.", "main_findings": "Separate creative prototyping from production acceptance; Add security and regression checks early; Train users to recognize the limits of superficial functional demos.", "bibliography_year": "2025", "section_id": "5.1", "chapter_id": "C05", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W045;W046;W050", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS039", "stable_bibliography_id": "O039", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Vulnerability and secure code", "K:Technical debt and complexity", "K:Human cognitive load" ], "volumes": [] }, { "id": "O040", "label": "MCP: What It Is and Why It Matters", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C06" ], "printed": false, "summary": "The Model Context Protocol standardizes agent access to external resources and tools, increasing integration reach and the importance of permission controls.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/mcp-what-it-is-and-why-it-matters", "authors": "Addy Osmani", "year": "2025-03-27", "attributes": { "label": "MCP: What It Is and Why It Matters", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/mcp-what-it-is-and-why-it-matters", "source_url": "https://addyo.substack.com/p/mcp-what-it-is-and-why-it-matters", "url": "https://addyo.substack.com/p/mcp-what-it-is-and-why-it-matters", "publication_date": "2025-03-27", "contribution": "The Model Context Protocol standardizes agent access to external resources and tools, increasing integration reach and the importance of permission controls.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Teach interface contracts separately from access policy; Review external tool descriptions as untrusted input; Constrain secrets, scopes and side-effecting calls.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "An accessible explanation of MCP clients, servers and tool interoperability, including opportunities for richer context. For the book it illustrates the distinction between a convenient integration protocol and the authorization, sandbox and trust decisions the protocol cannot solve alone.", "in_short": "The Model Context Protocol standardizes agent access to external resources and tools, increasing integration reach and the importance of permission controls.", "main_findings": "Teach interface contracts separately from access policy; Review external tool descriptions as untrusted input; Constrain secrets, scopes and side-effecting calls.", "bibliography_year": "2025", "section_id": "6.7", "chapter_id": "C06", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W053;W060", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "Package hallucination (W043) does not meaningfully address MCP tool authorization; security/harness links are more relevant.", "prior_record_id": "AOS040", "stable_bibliography_id": "O040", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Dependencies and agent supply chain", "K:Prompt injection and malicious tools", "K:Agent context and persistent memory" ], "volumes": [] }, { "id": "O041", "label": "Leading Effective Engineering Teams in the Age of GenAI", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C11" ], "printed": false, "summary": "Engineering leaders should define outcome quality, train teams and make human verification a normal part of AI adoption.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "url": "https://addyo.substack.com/p/leading-effective-engineering-teams-c9b", "authors": "Addy Osmani", "year": "2025-03-19", "attributes": { "label": "Leading Effective Engineering Teams in the Age of GenAI", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/leading-effective-engineering-teams-c9b", "source_url": "https://addyo.substack.com/p/leading-effective-engineering-teams-c9b", "url": "https://addyo.substack.com/p/leading-effective-engineering-teams-c9b", "publication_date": "2025-03-19", "contribution": "Engineering leaders should define outcome quality, train teams and make human verification a normal part of AI adoption.", "limitation": "Practitioner interpretation; not primary experimental evidence.", "manual_use": "Publish an explicit AI contribution policy; Allocate capacity for capability development and review; Evaluate delivery quality and learning, not generated-line counts.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani translates AI tool introduction into management duties: agree on what \"better\" means, avoid dependency on apparent speed, establish review expectations and support developer upskilling. It functions as organizational guidance rather than measured intervention.", "in_short": "Engineering leaders should define outcome quality, train teams and make human verification a normal part of AI adoption.", "main_findings": "Publish an explicit AI contribution policy; Allocate capacity for capability development and review; Evaluate delivery quality and learning, not generated-line counts.", "bibliography_year": "2025", "section_id": "11.3", "chapter_id": "C11", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W147;W149;W150", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner interpretation; not primary experimental evidence.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner interpretation; not primary experimental evidence.", "editorial_revision_detail": "", "prior_record_id": "AOS041", "stable_bibliography_id": "O041", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Human-agent collaboration", "K:Skills formation and teaching", "K:Specifications and governance" ], "volumes": [] }, { "id": "O042", "label": "Beyond the 70%: Maximizing the human 30% of AI-assisted coding", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:Productivity" ], "actual": [], "mapped": [ "C:C10" ], "printed": false, "summary": "The difficult last-mile work in AI-assisted development remains architecture, requirements, integration and verification.", "limitation": "The 70/30 split is a practitioner metaphor and is not independently measured in this essay.", "url": "https://addyo.substack.com/p/beyond-the-70-maximizing-the-human", "authors": "Addy Osmani", "year": "2025-03-13", "attributes": { "label": "Beyond the 70%: Maximizing the human 30% of AI-assisted coding", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/beyond-the-70-maximizing-the-human", "source_url": "https://addyo.substack.com/p/beyond-the-70-maximizing-the-human", "url": "https://addyo.substack.com/p/beyond-the-70-maximizing-the-human", "publication_date": "2025-03-13", "contribution": "The difficult last-mile work in AI-assisted development remains architecture, requirements, integration and verification.", "limitation": "The 70/30 split is a practitioner metaphor and is not independently measured in this essay.", "manual_use": "Distinguish generated scaffold from production-ready change; Teach system-level debugging and design; Avoid quoting 70/30 as a measured software engineering constant.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Expanding his 2024 \"70% problem\" framing, Osmani discusses the durable value of engineers who can interpret behavior, handle edge cases and maintain systems. The percentages are rhetorical heuristics for relative task difficulty, not tested ratios of project completion.", "in_short": "The difficult last-mile work in AI-assisted development remains architecture, requirements, integration and verification.", "main_findings": "Distinguish generated scaffold from production-ready change; Teach system-level debugging and design; Avoid quoting 70/30 as a measured software engineering constant.", "bibliography_year": "2025", "section_id": "10.9", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W117;W154;W010", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "The 70/30 split is a practitioner metaphor and is not independently measured in this essay.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "The 70/30 split is a practitioner metaphor and is not independently measured in this essay.", "editorial_revision_detail": "The article concerns human competency and use of the non-automated portion of software work; a capability-development contract is the better teaching placement.", "prior_record_id": "AOS042", "stable_bibliography_id": "O042", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:System understanding and rationale", "K:Skills formation and teaching", "K:Output versus shipped value" ], "volumes": [] }, { "id": "O043", "label": "Applied \"Software Engineering at Google\"", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C07" ], "printed": false, "summary": "Testing, review, documentation and sustainable ownership are established software-engineering practices that remain useful with AI-generated code.", "limitation": "Expository synthesis of an existing software engineering book, not original AI research.", "url": "https://addyo.substack.com/p/applied-software-engineering-at-google", "authors": "Addy Osmani", "year": "2025-03-06", "attributes": { "label": "Applied \"Software Engineering at Google\"", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/applied-software-engineering-at-google", "source_url": "https://addyo.substack.com/p/applied-software-engineering-at-google", "url": "https://addyo.substack.com/p/applied-software-engineering-at-google", "publication_date": "2025-03-06", "contribution": "Testing, review, documentation and sustainable ownership are established software-engineering practices that remain useful with AI-generated code.", "limitation": "Expository synthesis of an existing software engineering book, not original AI research.", "manual_use": "Reuse established tests-and-change reasoning; Distinguish well-founded engineering precedent from measured agent performance; Use as contextual support for governance.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani adapts lessons from the public Software Engineering at Google book, including tests that make change safer, code review and organizational learning. It supplies historical engineering principles, not direct evidence that an AI technique is effective.", "in_short": "Testing, review, documentation and sustainable ownership are established software-engineering practices that remain useful with AI-generated code.", "main_findings": "Reuse established tests-and-change reasoning; Distinguish well-founded engineering precedent from measured agent performance; Use as contextual support for governance.", "bibliography_year": "2025", "section_id": "7.4", "chapter_id": "C07", "relevance": "Adjacent (general engineering)", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W064;W161;W173", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Expository synthesis of an existing software engineering book, not original AI research. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONTEXTUAL_ONLY", "article_is_ai_specific": "false", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Expository synthesis of an existing software engineering book, not original AI research. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "editorial_revision_detail": "Conspicuously prevent non-AI engineering essays from being reported as evidence of AI-specific outcomes.", "prior_record_id": "AOS043", "stable_bibliography_id": "O043", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Independent verification and oracles", "K:Architecture and modularity", "K:Ownership and knowledge transfer" ], "volumes": [] }, { "id": "O044", "label": "Speech-to-Code: Vibe Coding with Voice", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Productivity" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "Voice input changes how developers express tasks to coding assistants but does not remove the need to verify resulting software.", "limitation": "First-person tool demonstration without measured comparative productivity.", "url": "https://addyo.substack.com/p/speech-to-code-vibe-coding-with-voice", "authors": "Addy Osmani", "year": "2025-02-28", "attributes": { "label": "Speech-to-Code: Vibe Coding with Voice", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/speech-to-code-vibe-coding-with-voice", "source_url": "https://addyo.substack.com/p/speech-to-code-vibe-coding-with-voice", "url": "https://addyo.substack.com/p/speech-to-code-vibe-coding-with-voice", "publication_date": "2025-02-28", "contribution": "Voice input changes how developers express tasks to coding assistants but does not remove the need to verify resulting software.", "limitation": "First-person tool demonstration without measured comparative productivity.", "manual_use": "Teach modality-independent task specification; Evaluate accessibility and ergonomics separately from correctness; Do not extrapolate a personal speedup to all teams.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani demonstrates dictating instructions into an AI coding workflow using speech tools and describes personal ergonomic and velocity benefits. The improvement is experiential; no controlled comparison of input methods or final code quality is provided.", "in_short": "Voice input changes how developers express tasks to coding assistants but does not remove the need to verify resulting software.", "main_findings": "Teach modality-independent task specification; Evaluate accessibility and ergonomics separately from correctness; Do not extrapolate a personal speedup to all teams.", "bibliography_year": "2025", "section_id": "1.2", "chapter_id": "C01", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "First-person tool demonstration without measured comparative productivity.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "First-person tool demonstration without measured comparative productivity.", "editorial_revision_detail": "Autocomplete acceptance and generic self-reported AI productivity are not evidence of voice-to-code efficacy; retain one workflow-context comparison.", "prior_record_id": "AOS044", "stable_bibliography_id": "O044", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Human cognitive load", "K:Human-agent collaboration" ], "volumes": [] }, { "id": "O045", "label": "Personal Software: The Unbundling of the Programmer?", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Economics", "T:Jobs & roles" ], "actual": [], "mapped": [ "C:C12" ], "printed": false, "summary": "Lower creation costs may expand demand for one-off and personal applications, an economic hypothesis distinct from measured productivity or employment effects.", "limitation": "Forward-looking economic scenario; no primary adoption or welfare evaluation.", "url": "https://addyo.substack.com/p/personal-software-the-unbundling", "authors": "Addy Osmani", "year": "2025-02-09", "attributes": { "label": "Personal Software: The Unbundling of the Programmer?", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Economics", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/personal-software-the-unbundling", "source_url": "https://addyo.substack.com/p/personal-software-the-unbundling", "url": "https://addyo.substack.com/p/personal-software-the-unbundling", "publication_date": "2025-02-09", "contribution": "Lower creation costs may expand demand for one-off and personal applications, an economic hypothesis distinct from measured productivity or employment effects.", "limitation": "Forward-looking economic scenario; no primary adoption or welfare evaluation.", "manual_use": "Separate prototype creation cost from lifecycle cost; Evaluate new demand and substitution scenarios independently; Mark market forecasts as hypotheses.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani explores a world in which individuals and domain specialists build small personalized applications too narrow to justify traditional development contracts. The claim is about enabling software that previously would not be built at all. It is a forward-looking long-tail demand scenario, not a measured labor substitution, return-on-investment or full lifecycle cost study.", "in_short": "Lower creation costs may expand demand for one-off and personal applications, an economic hypothesis distinct from measured productivity or employment effects.", "main_findings": "Separate prototype creation cost from lifecycle cost; Evaluate new demand and substitution scenarios independently; Mark market forecasts as hypotheses.", "bibliography_year": "2025", "section_id": "12.8", "chapter_id": "C12", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W145", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Forward-looking economic scenario; no primary adoption or welfare evaluation. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Forward-looking economic scenario; no primary adoption or welfare evaluation. Framework or market hypothesis is an authored practitioner proposal, not an empirically validated scale or intervention.", "editorial_revision_detail": "The central proposition is a distributional/demand shift toward personally created software, not choosing an autonomy level by least cost.; Firm-level AI hiring studies do not investigate the novel long-tail personal-software demand hypothesized by the article; retain the lifecycle ROI paper only as economic contrast.; State the original essay’s demand-creation thesis explicitly.; Emphasize personalization and software demand rather than autonomy-cascade cost optimization.; The paper is a qualitative demand scenario; employment comparison is contextual rather than observed outcome.; Do not promote conceptual taxonomy or market hypotheses into established research outcomes.", "prior_record_id": "AOS045", "stable_bibliography_id": "O045", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Economic costs and return", "K:Developer workflow and adoption", "K:Employment and early-career roles" ], "volumes": [] }, { "id": "O046", "label": "Why I use Cline for AI Engineering", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "Cline illustrates how a tool-using coding assistant differs from a conventional completion engine through its context and environment access.", "limitation": "Subjective tool review, updated 2025-07-21; product comparisons are not controlled benchmarks.", "url": "https://addyo.substack.com/p/why-i-use-cline-for-ai-engineering", "authors": "Addy Osmani", "year": "2025-01-30", "attributes": { "label": "Why I use Cline for AI Engineering", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/why-i-use-cline-for-ai-engineering", "source_url": "https://addyo.substack.com/p/why-i-use-cline-for-ai-engineering", "url": "https://addyo.substack.com/p/why-i-use-cline-for-ai-engineering", "publication_date": "2025-01-30", "contribution": "Cline illustrates how a tool-using coding assistant differs from a conventional completion engine through its context and environment access.", "limitation": "Subjective tool review, updated 2025-07-21; product comparisons are not controlled benchmarks.", "manual_use": "Classify assistant capabilities by observable tools and permissions; Evaluate cost and context visibility on representative tasks; Treat personal vendor preferences as dated, not benchmark evidence.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani offers a first-person comparison of coding assistants, favoring Cline for file, terminal, browser and debugging workflows while acknowledging trade-offs. The post was updated in July 2025; its feature and product assessments are time-specific.", "in_short": "Cline illustrates how a tool-using coding assistant differs from a conventional completion engine through its context and environment access.", "main_findings": "Classify assistant capabilities by observable tools and permissions; Evaluate cost and context visibility on representative tasks; Treat personal vendor preferences as dated, not benchmark evidence.", "bibliography_year": "2025", "section_id": "3.2", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002;W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Subjective tool review, updated 2025-07-21; product comparisons are not controlled benchmarks.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Subjective tool review, updated 2025-07-21; product comparisons are not controlled benchmarks.", "editorial_revision_detail": "Time-specific Cline feature review is not validated by a later Microsoft adoption study (W014) or a virtual memory system (W157).", "prior_record_id": "AOS046", "stable_bibliography_id": "O046", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Agent context and persistent memory", "K:Agent failure and escalation" ], "volumes": [] }, { "id": "O047", "label": "AI-Driven Prototyping: v0, Bolt, and Lovable Compared", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03" ], "printed": false, "summary": "AI prototyping services reduce time from concept to demonstrable app but vary in architecture, control and production suitability.", "limitation": "Practitioner tool comparison; tool features have evolved since early 2025.", "url": "https://addyo.substack.com/p/ai-driven-prototyping-v0-bolt-and", "authors": "Addy Osmani", "year": "2025-01-11", "attributes": { "label": "AI-Driven Prototyping: v0, Bolt, and Lovable Compared", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "What agents can do", "year": "2025", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/ai-driven-prototyping-v0-bolt-and", "source_url": "https://addyo.substack.com/p/ai-driven-prototyping-v0-bolt-and", "url": "https://addyo.substack.com/p/ai-driven-prototyping-v0-bolt-and", "publication_date": "2025-01-11", "contribution": "AI prototyping services reduce time from concept to demonstrable app but vary in architecture, control and production suitability.", "limitation": "Practitioner tool comparison; tool features have evolved since early 2025.", "manual_use": "Choose tools by intended artifact and handoff needs; Budget refactoring and testing after a prototype; Do not treat a point-in-time vendor comparison as a reproducible benchmark.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "The essay compares v0, Bolt and Lovable for common proof-of-concept workflows, noting different strengths in UI scaffolding and fuller application assembly. It distinguishes product exploration from subsequent testing, maintainability and integration work.", "in_short": "AI prototyping services reduce time from concept to demonstrable app but vary in architecture, control and production suitability.", "main_findings": "Choose tools by intended artifact and handoff needs; Budget refactoring and testing after a prototype; Do not treat a point-in-time vendor comparison as a reproducible benchmark.", "bibliography_year": "2025", "section_id": "3.3", "chapter_id": "C03", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W002;W191", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Practitioner tool comparison; tool features have evolved since early 2025.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2025", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "REVISED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Practitioner tool comparison; tool features have evolved since early 2025.", "editorial_revision_detail": "A personal comparison of three prototyping tools is not a formal SWE-agent benchmark; use as an example in task envelopes/capability fit.; MirrorCode (W193) measures extended coding tasks, not vendors v0/Bolt/Lovable. Retain two broad agent workflow comparisons only.", "prior_record_id": "AOS047", "stable_bibliography_id": "O047", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Developer workflow and adoption", "K:Architecture and modularity", "K:Specifications and governance" ], "volumes": [] }, { "id": "O048", "label": "Future-proofing your Software Engineering career", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Skills & learning", "T:Jobs & roles" ], "actual": [], "mapped": [ "C:C10" ], "printed": false, "summary": "Career resilience in AI-assisted software engineering rests on adaptable domain expertise, systems judgment and learning.", "limitation": "Career advice and forecasts, not original labor-market empirical research.", "url": "https://addyo.substack.com/p/future-proofing-your-software-engineering", "authors": "Addy Osmani", "year": "2024-12-23", "attributes": { "label": "Future-proofing your Software Engineering career", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Skills & learning", "year": "2024", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/future-proofing-your-software-engineering", "source_url": "https://addyo.substack.com/p/future-proofing-your-software-engineering", "url": "https://addyo.substack.com/p/future-proofing-your-software-engineering", "publication_date": "2024-12-23", "contribution": "Career resilience in AI-assisted software engineering rests on adaptable domain expertise, systems judgment and learning.", "limitation": "Career advice and forecasts, not original labor-market empirical research.", "manual_use": "Build a transferable engineering-skills portfolio; Practice validating and integrating AI work; Separate career advice from measured employment effects.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani surveys the tasks models can automate and emphasizes software architecture, evaluation, product understanding and the ability to learn new techniques. Predictions about changing roles are presented as guidance, not labor-market identification.", "in_short": "Career resilience in AI-assisted software engineering rests on adaptable domain expertise, systems judgment and learning.", "main_findings": "Build a transferable engineering-skills portfolio; Practice validating and integrating AI work; Separate career advice from measured employment effects.", "bibliography_year": "2024", "section_id": "10.9", "chapter_id": "C10", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W118;W122;W135", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "Career advice and forecasts, not original labor-market empirical research.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2024", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "Career advice and forecasts, not original labor-market empirical research.", "editorial_revision_detail": "", "prior_record_id": "AOS048", "stable_bibliography_id": "O048", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Skills formation and teaching", "K:Employment and early-career roles", "K:System understanding and rationale" ], "volumes": [] }, { "id": "O049", "label": "The 70% problem: Hard truths about AI-assisted coding", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Productivity", "T:Code quality & security" ], "actual": [ "C:C02" ], "mapped": [ "C:C02" ], "printed": true, "summary": "Getting a prototype mostly implemented with AI is far easier than completing the remaining integrations and engineering assurances.", "limitation": "The 70% estimate is an anecdotal framing, not a validated quantitative benchmark.", "url": "https://addyo.substack.com/p/the-70-problem-hard-truths-about", "authors": "Addy Osmani", "year": "2024-12-04", "attributes": { "label": "The 70% problem: Hard truths about AI-assisted coding", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Productivity", "year": "2024", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/the-70-problem-hard-truths-about", "source_url": "https://addyo.substack.com/p/the-70-problem-hard-truths-about", "url": "https://addyo.substack.com/p/the-70-problem-hard-truths-about", "publication_date": "2024-12-04", "contribution": "Getting a prototype mostly implemented with AI is far easier than completing the remaining integrations and engineering assurances.", "limitation": "The 70% estimate is an anecdotal framing, not a validated quantitative benchmark.", "manual_use": "Measure end-to-end shipping and rework instead of scaffolding speed; Identify integration and maintenance as separate work; Preserve the heuristic character of the percentage.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani contrasts fast \"zero-to-MVP\" generation with difficult iteration, debugging and production readiness in realistic codebases. His memorable 70% phrasing is descriptive and experiential, not a controlled measurement of task success.", "in_short": "Getting a prototype mostly implemented with AI is far easier than completing the remaining integrations and engineering assurances.", "main_findings": "Measure end-to-end shipping and rework instead of scaffolding speed; Identify integration and maintenance as separate work; Preserve the heuristic character of the percentage.", "bibliography_year": "2024", "section_id": "2.5", "chapter_id": "C02", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "W004;W010;W055", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "The 70% estimate is an anecdotal framing, not a validated quantitative benchmark.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2024", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONFIRMED", "article_is_ai_specific": "true", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "The 70% estimate is an anecdotal framing, not a validated quantitative benchmark.", "editorial_revision_detail": "", "prior_record_id": "AOS049", "stable_bibliography_id": "O049", "edition_integration": "CORE_CITED", "manuscript_citation_status": "ACTUAL_CITATION", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "manuscript_section_id": "2.5", "manuscript_citation_basis": "LaTeX \\osmanicite parsed within containing \\section, Oct 10 edition integrity review", "ndse2_actual_chapters_json": "[\"C:C02\"]", "ndse2_in_bibliography": true, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Output versus shipped value", "K:Technical debt and complexity", "K:Independent verification and oracles" ], "volumes": [ "V06" ] }, { "id": "O050", "label": "User-centric engineering.", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Engineering structure" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "Customer needs and observable outcomes—not lines of code—should determine whether an engineering change succeeds.", "limitation": "General user-experience and product principles, no original AI evaluation.", "url": "https://addyo.substack.com/p/user-centric-engineering", "authors": "Addy Osmani", "year": "2024-11-15", "attributes": { "label": "User-centric engineering.", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Engineering structure", "year": "2024", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/user-centric-engineering", "source_url": "https://addyo.substack.com/p/user-centric-engineering", "url": "https://addyo.substack.com/p/user-centric-engineering", "publication_date": "2024-11-15", "contribution": "Customer needs and observable outcomes—not lines of code—should determine whether an engineering change succeeds.", "limitation": "General user-experience and product principles, no original AI evaluation.", "manual_use": "Start with a concrete user problem and a user-observable success criterion; Validate design trade-offs through user interviews, usage data, and support feedback; Use only as background product engineering context, not as evidence of AI productivity.", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "B", "full_manuscript_audited": "article webpage read; linked empirical sources not all independently audited", "summary": "Osmani argues that engineering decisions should start with user needs and observable benefit rather than technical elegance. He offers questions to ask before development, during iterative implementation and after launch, supported by practices such as support-ticket analysis, behavioral observation, interviews and product feedback. The article does not evaluate AI coding systems.", "in_short": "Customer needs and observable outcomes—not lines of code—should determine whether an engineering change succeeds.", "main_findings": "Start with a concrete user problem and a user-observable success criterion; Validate design trade-offs through user interviews, usage data, and support feedback; Use only as background product engineering context, not as evidence of AI productivity.", "bibliography_year": "2024", "section_id": "1.1", "chapter_id": "C01", "relevance": "Adjacent (general engineering)", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "related_works": "", "provenance": "Osmani public Substack sitemaps 2024/2025/2026 plus individual source URLs; independent editorial review 2026-10-10", "baseline_provenance": "Overlay only; does not change any original W001-W193 evidence classification", "editorial_caution": "General user-experience and product principles, no original AI evaluation. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/sitemap/2024", "publication_status_note": "Substack essay, not a peer-reviewed research paper", "review_status": "CONTEXTUAL_ONLY", "article_is_ai_specific": "false", "review_date": "2026-10-10", "research_link_semantics": "comparison/analogy, not observed citation, causal validation or replication", "audit_evidence_note": "General user-experience and product principles, no original AI evaluation. CONTEXTUAL ONLY: this article does not evaluate AI-assisted development, agent behavior, or causal AI outcomes.", "editorial_revision_detail": "Product empathy is general engineering context; productivity effects of AI are not assessed.; Drop unwarranted human-agent collaboration topic from pre-AI conceptual essay.; A user-centric product essay contains no AI outcome evidence; remove generic AI productivity study links.; Preserve the source’s actual user-empathy framework without writing agent-era conclusions into the author’s summary.; Avoid attributing a prompt-generation workflow to a general software product essay.; Conspicuously prevent non-AI engineering essays from being reported as evidence of AI-specific outcomes.", "prior_record_id": "AOS050", "stable_bibliography_id": "O050", "edition_integration": "CATALOGUE_ONLY", "manuscript_citation_status": "NOT_CITED_IN_MAIN_TEXT", "provenance_base": "Oct 10 2026 694-node graph (not 9 Oct graph)", "evidence_is_experimental": "false", "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations" }, "topics": [ "K:Specifications and governance", "K:Output versus shipped value" ], "volumes": [] }, { "id": "O051", "label": "Agentic Verification", "kind": "practitioner_article", "collection": "O", "category": null, "streams": [ "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C05", "C:C07", "C:C08" ], "printed": false, "summary": "Reliable agent work depends on executable checks: realistic user flows, explicit properties and differential comparisons, with fast feedback and human-owned success criteria.", "limitation": "Practitioner guidance, not a controlled experiment or proof of universal correctness; generated tests and oracle independence still need scrutiny.", "url": "https://addyo.substack.com/p/agentic-verification", "authors": "Addy Osmani", "year": "2026-10-14", "attributes": { "label": "Agentic Verification", "kind": "practitioner_article", "node_class": "CURATED_PRACTITIONER_NOT_RESEARCH", "group": "Verification & review", "year": "2026", "lead_author": "Addy Osmani", "source": "https://addyo.substack.com/p/agentic-verification", "source_url": "https://addyo.substack.com/p/agentic-verification", "url": "https://addyo.substack.com/p/agentic-verification", "publication_date": "2026-10-14", "evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "reviewed_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "original_evidence_class": "Practitioner essay / technical guidance (not peer-reviewed)", "editorial_priority": "A", "full_manuscript_audited": "Full primary article read; no statistical replication performed", "summary": "Osmani recommends autonomous feedback loops using end-to-end checks, property-based testing and old/new behavioral comparisons. Checks should run quickly and deterministically, preserve actionable failures, and be grounded in success criteria the human owns.", "manual_use": "Define the oracle before implementation; retain failing inputs; inspect generated tests; run heavier integration checks before release.", "bibliography_year": "2026", "chapter_id": "C07", "section_id": "", "relevance": "Direct", "method_status": "curated practitioner article (not a research study)", "primary_research": "false", "provenance": "Elevate public archive and full primary article inspected 2026-10-18", "baseline_provenance": "Incremental addition to frozen v12.1 catalogue; O001-O050 unchanged", "original_url_verified": "yes", "publication_source_url": "https://addyo.substack.com/archive", "publication_status_note": "Published practitioner article; online companion update", "review_status": "FULL_TEXT_REVIEWED", "article_is_ai_specific": "true", "review_date": "2026-10-18", "stable_bibliography_id": "O051", "edition_integration": "ONLINE_COMPANION_UPDATE", "manuscript_citation_status": "NOT_CITED_IN_PUBLISHED_NDSE2_13_7_1", "evidence_is_experimental": "false", "source_access": "Full public text", "research_link_semantics": "Editorial thematic mapping; no assertion of literal citation, causal validation or replication", "limitation": "Practitioner guidance, not a controlled experiment or proof of universal correctness; generated tests and oracle independence still need scrutiny.", "manuscript_citation_basis": "Absent from the final published NDSE2 13.7.1 active EN/ITA source tree and loaded bibliography", "atlas_added_on": "2026-10-18", "canonical_url": "https://addyo.substack.com/p/agentic-verification", "online_companion_update": true, "ndse2_actual_chapters_json": "[]", "ndse2_in_bibliography": false, "ndse2_citation_basis": "Final 13.7.1 active EN/ITA source tree; index links and nocite are not chapter citations", "pre_reconciliation_manuscript_citation_status": "NOT_ESTABLISHED_FROM_PUBLISHED_MANUSCRIPT", "pre_reconciliation_manuscript_citation_basis": "No published-manuscript citation inferred by this source review; reconcile against final NDSE2 sources." }, "topics": [ "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:Specifications and governance", "K:Long-running agent execution" ], "volumes": [], "pre_reconciliation_mapped": [ "C:C07", "C:C04", "C:C05", "C:C08" ] } ] }