Agents · Case studiesThe reviewer in the roomA human looked at what the agent produced and said no. That's not a training-wheels phase. It's how the agent learns what accurate means here.James Dodd·16 Apr 2026·4 min read
Agents · Case studiesWe built one AI to commit fraud and another to stop itBoth agents found a way. Neither was broken. That was the problem, and it taught us where AI earns its keep and where it doesn't.James Dodd·17 Mar 2026·7 min read