I use AI agents every day: not just to write code, but for research, brainstorming, planning, and user testing. I wrote up how I code with AI agents in a longer article.
It covers working spec first, quality gates on every commit, worktrees, parallel agents, and throwing away code that misses.
The short version:
- Write the spec first, then build from a plan the agent and I both reviewed.
- Run quality gates on every commit so the agent catches its own mistakes.
- Give each branch its own worktree and environment.
- Run agents in parallel on separate tasks.
- Throw away code that misses. It’s cheap to regenerate.
