FIELD NOTES / DEVELOPING IN PUBLIC
Building software
with AI agents.
Practical notes for engineers and tech leads making coding agents work on a real team: playbooks, evidence, and checklists you can use this week.
WRITE / BUILD / REVISIT
What I Write About
Three pillars, one reader: the engineer who has to make agents work in production.
AI-assisted engineering
Agent setup, AGENTS.md, reviewing AI-generated pull requests, testing, and cost.
Agents in production
Harnesses, memory, autonomy, and agents running real operations, including game LiveOps.
Engineering craft
Team process, product thinking, and the modern web stack behind this site.
RECENTLY PUBLISHED
Latest Posts
Harness Engineering for Coding Agents: A Practical Team Setup
Same model, different results: the harness decides. What a team controls around Claude Code or Codex, with hook configs, fast checks, and a setup checklist.
How to Review AI-Generated Pull Requests: A Risk-First Checklist
Agents now start over 40% of GitHub pull requests, and review is the bottleneck. A risk-tiered process, an evidence-first checklist, and a PR template to adopt.
Does AGENTS.md Help Coding Agents? What 2026 Research Actually Shows
Three 2026 studies disagree on whether AGENTS.md helps coding agents. What the evidence supports, what to cut, and a one-afternoon eval to test your file.