Updates
What's New on AIDLC.io
Every page on the site, newest change first. The field this guide covers moves quickly, so pages are re-researched on a regular schedule and dated when they change - this list is built from those dates, so it is always current.
How the Site Stays Current
A twice-monthly review
Every page is re-checked twice a month against the current state of the tools, standards and research it describes. Stale product names, dates and recommendations are corrected, and pages are added where the field has moved.
Sources, or it does not ship
Any figure on the site names its source in the same sentence, and the Research Library lists every one. A claim that cannot be traced to its source is corrected or removed rather than left standing.
Dated where it matters
Pages about fast-moving tools and vendor features carry a visible "Last reviewed" line, and every article carries its modified date in its structured data. The list below is built from those same dates.
Recently Updated Pages
Grouped by the date each page last changed. Prefer a feed reader? The same list is published as an Atom feed.
(38 pages)
A Worked Example of AIDLCFollow one invented feature from a customer request to a production incident, with the actual artifact each AIDLC phase hands to the next one.Agent Orchestration: The Control Plane Is the ProductHow to conduct a fleet of AI agents across the SDLC: a state machine that owns every transition, dual review, and fail-closed gates you can prove.AI Agents and Tools for AIDLCA tool-agnostic map of the 2026 AI landscape - coding agents, MCP, orchestration, and observability - across all five phases of the AIDLC framework.AI Code Review at Scale: What the Data ShowsAI code review lifts measured quality while human oversight quietly erodes. See the 2025-2026 data, review standards, fatigue fixes, and merge ownership rules.AI Governance Before Scale: 5 PillarsGovernance before rollout: five pillars, a policy outline, and vendor checklists for teams giving AI coding agents real access to code.AI SDLC Security and ComplianceThe AI SDLC threat model, data classification tiers, and how SOC 2, HIPAA, ISO 27001, NIST, and the EU AI Act map to it.AI SDLC Templates and ChecklistsCopy-paste templates for AI usage policy, ticket contracts, PR fields, reviewer checklists, agent hard rules, and vendor and incident checklists.AIDLC Analyze Phase - AI-Assisted RequirementsDeep dive into the Analyze phase of AIDLC and see how AI assists with requirements gathering, research synthesis, and feasibility analysis.AIDLC Cheatsheet - Quick Reference GuideQuick reference cheatsheet for AIDLC methodology. Instant access to all five phases, key practices, tools, and implementation tips for AI-Native development.AIDLC Curate Phase - AI-Powered MonitoringExplore the Curate phase of AIDLC. Learn about AI-powered monitoring, intelligent debugging, continuous documentation, and iterative improvement.AIDLC Develop Phase - AI Coding AgentsDeep dive into the Develop phase of AIDLC. Learn how AI coding agents plan and execute multi-step tasks while you scope, steer, and review the diffs.AIDLC GlossaryPlain-English definitions of the terms that matter in AI-native software delivery - agents, harnesses, context engineering, MCP, and spec-driven development.AIDLC Ideate Phase - AI-Driven ArchitectureMaster the Ideate phase of AIDLC. Explore AI-driven architecture design, documentation generation, and solution modeling before writing code.AIDLC Launch Phase - AI-Powered CI/CDMaster the Launch phase of AIDLC. Learn about AI-generated tests, intelligent CI/CD configuration, deployment automation, and release validation.AIDLC on Legacy CodeHow to use AI agents on legacy code safely: build context, pin behavior with characterization tests, strangle in slices, and upgrade in small batches.AIDLC Prompt LibraryA tool-agnostic library of prompts for every AIDLC phase - copy, adapt, and use them to scope work, steer agents, and review what comes back.AIDLC vs AWS AI-DLCAIDLC and AWS AI-DLC share a name but not a shape. Compare the five-phase methodology with AWS's 33-stage workflow engine and see when to use each.AIDLC vs the ML LifecycleAI development life cycle means two things: building software with AI agents (AIDLC) or building a model (MLOps). See the table, standards, and where they meet.AIDLC With Scrum and KanbanRun AIDLC inside Scrum or Kanban: map the five phases to sprint events, size work by review effort, cap review WIP, and define Done for agent-authored code.AIDLC.io Privacy PolicyAIDLC.io collects anonymous usage statistics and what you send through the contact form. What is collected, why, for how long, and how to opt out.Context and Spec-Driven DevelopmentContext engineering and spec-driven development are the core skill of AI-native delivery: how to write specs, steer agents, and review their output well.Engineering Practices That Make AI WorkEleven engineering practices for AI-assisted delivery: context files, ticket contracts, plan-first review, mutation testing, and risk-tiered review.Field Notes - Building With AI AgentsFirst-hand notes from two production AI systems: the language drift that cost weeks, why a non-frontier model won, and the habit that changed everything.How to Choose an AI Coding AgentA framework for choosing an AI coding agent - the axes that actually matter, how to run a fair trial on real work, and the mistakes to avoid.Learn AIDLC - Complete Training GuideMaster the AIDLC framework with comprehensive training materials, examples, and best practices for implementing AI-Native Software Delivery.Measuring AI Impact the Right WayWhich pre-AI baseline to capture, which metrics to report, which vanity numbers to skip, and the quarterly test that flags a queue, not a gain.Running Agents in the CI PipelineHow to run coding agents in CI safely: service identity, scoped permissions, sandboxing, budgets, and seven hard rules enforced in the workflow.Teams and Skills in an AI-Native SDLCHow roles shift when agents write first drafts: junior skill formation, interviewing, upskilling, review fatigue, and measuring developer experience.The AI Engineering Cost Model, ExplainedHow seats, seat allowances, and API usage make up AI engineering spend, plus the four-position build-vs-buy framework for agent tooling.The AI Productivity Paradox, ExplainedFaros AI, DORA, and METR data on the AI productivity paradox: individual output rises while organizational throughput and stability often do not.The AI SDLC Maturity Model: Levels 0-4Score your team against five AI SDLC maturity levels, from no AI to governed and optimized, with a 12-question interactive self-assessment.The AI SDLC Playbook: Ten Rules, Three ErasThe hub for AI in the software development lifecycle: ten operating rules, three eras of AI, and where each phase gets help or fails.The AI SDLC Research Library and SourcesEvery study, survey, and standard the AI SDLC Playbook cites, with the key finding from each and a link to the primary source.The AI SDLC Rollout Playbook GuideA four-phase timeline for rolling out AI-assisted engineering: baseline in three weeks, assist, delegate, and compound - with exit criteria for each phase.The AI-Native SDLC: Seven PillarsThe seven pillars of an AI-native SDLC: context and measurement as platform investments, plus five workflow shifts from planning to maintenance.The Hidden Risks of AI-Generated CodeThe 13-risk register for AI-assisted development, backed by CodeRabbit, GitClear, Veracode, Apiiro, and developer trust survey data.The State of AI in 2026An autumn 2026 snapshot of AI development: agents as the default, portable agent standards, and the discipline of verifying work you did not write.Twelve Ways AI SDLC Adoption FailsThe same twelve failure patterns recur across AI adoptions of every size - each one has a name, an observable tell, and a documented fix.
Spotted Something Out of Date?
Corrections and better sources are welcome. Send the page and what changed, and the fix goes into the next review.