Agentic AI
3 articles on this topic.
- I don't just review the agent's code anymore. I ask for proofSix Playwright flows that leave a screenshot, video and trace on failure. In CI on every PR and every deploy. And four times the evidence proved me wrong.
- Nobody reviews their own code well, not even an agentThree bugs that passed tests and review without looking odd. A fresh reviewer, human or agent, won't see them without knowing something the author didn't.
- My subagents reminded me of Naruto's shadow clonesI built eight subagents across two repos. In 18 hours they ran 27 times, 11 as real agents: what I delegate, what they found and what I still don't know.