Posts

An AI agent can change few files, run few tests, and say ‘done’ while the repository is still broken. A Task Completion Protocol in AGENTS.md defines what ‘done’ actually means.

Commit messages become write-only noise exactly when you need them most. These two small AI prompts turn staged diffs into useful commits, then turn those commits into a better pull request.

A structured diagnostic loop for prompt tuning inside the working context. Learn how to pivot AI agents from doing to reporting for better reasoning analysis.

Most backend architectures optimize for theoretical purity while ignoring real constraints. Learn how pragmatic layered architecture delivers 90% of Clean Architecture’s benefits with 10% of the ceremony—principles that work across any language or framework.