https://smcgowan.me/feed/ai-sdlc.xmlYou can't install an AI-SDLCAn article series by Stephen Mc Gowan on growing an AI-enabled software development lifecycle. 2026-09-24T10:02:16+10:00 Stephen Mc Gowan https://smcgowan.me/ Jekyll © 2026 Stephen Mc Gowan /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png Interlude: Jev and the economics of no-training2026-09-23T00:00:00+10:00 2026-09-23T00:00:00+10:00 https://smcgowan.me/posts/economics-of-no-training/ Stephen Mc Gowan I re-ran an old AWS Comprehend project against TypeSafe's Jev. With no training data it beat the trained classifier at roughly a two-hundredth of the cost. It also showed where the risk goes when you stop training models and start writing questions. Architecting for agent legibility2026-09-17T00:00:00+10:00 2026-09-17T00:00:00+10:00 https://smcgowan.me/posts/architecting-for-agent-legibility/ Stephen Mc Gowan An AI agent is a permanently new team member: it re-pays your codebase's discovery cost every single session. What that means for repo structure, docs as code, restraint about abstraction, and conventions machines can check. Build the quality floor before the autonomy2026-09-10T00:00:00+10:00 2026-09-10T00:00:00+10:00 https://smcgowan.me/posts/quality-floor-before-autonomy/ Stephen Mc Gowan Nineteen phantom products, the pull request that pivoted a proof of concept into a product, and what a ratcheting coverage floor actually buys when agents write most of the code: regression detection at machine speed, so the human can follow the bottleneck. Constrain, observe, review: agent safety as infrastructure2026-09-04T00:00:00+10:00 2026-09-04T00:00:00+10:00 https://smcgowan.me/posts/agent-safety-as-infrastructure/ Stephen Mc Gowan A self-hosted agent sandbox with exactly three egress paths, the breakout attempt it contained, and the reference architecture enterprise agent platforms should be heading for: every path an agent can take runs through a control point you own. Show me what you think we built2026-08-26T00:00:00+10:00 2026-08-26T00:00:00+10:00 https://smcgowan.me/posts/show-me-what-you-think-we-built/ Stephen Mc Gowan Generated customer updates, questionnaires with assumptions attached, and a solution design that audits its own citations: reading what the AI thinks you built is how you test architecture and finish the last 10 per cent. Engineered disagreement2026-08-20T00:00:00+10:00 2026-08-20T00:00:00+10:00 https://smcgowan.me/posts/engineered-disagreement/ Stephen Mc Gowan Four AI personas that changed nothing, one that changed everything, and what it actually takes to stop two agents on the same model from agreeing with each other. You can't install an AI-SDLC2026-08-11T00:00:00+10:00 2026-08-11T00:00:00+10:00 https://smcgowan.me/posts/you-cant-install-an-ai-sdlc/ Stephen Mc Gowan Why an AI-enabled software development lifecycle (SDLC) has to be grown, not installed: eighteen weeks of evidence from one human directing AI agents on a production system.