Tag · ai
#ai
3 posts tagged ai. View all posts.
AGENTS.md and CLAUDE.md are table stakes — enforcement is the moat
Every serious SaaS boilerplate now ships an AGENTS.md or a CLAUDE.md — a soft file the agent can quietly ignore by prompt 50. The differentiator is enforcement: ESLint-checked DDD boundaries and typed contracts that turn the build red when the agent crosses a line.
Why AI coding agents keep breaking things that already worked
AI generated code technical debt has a shape — agents drift toward their own defaults and quietly break code that already shipped. Why it happens, and what actually holds the line.
DX is a feature: how we decide what ships
Our internal heuristic for choosing libraries, abstractions, and patterns — biased toward what an LLM agent can navigate without help.