I've been thinking about this for a while and I'm sure other people have too. Maybe someone already solved it better than me. But here's the problem as I see it. Your coding agent can build a feature in 20 minutes. Tests pass, code is clean, the PR looks professional. You approve it. Three months later nobody uses it, or it works but solves a slightly different problem than the one you started with. The code was fine. The decision behind it wasn't. The real problem might be you sitting alone with your agent Code reviews catch bugs and patterns. They don't catch bad product decisions. By the time someone sees a diff, there's an unspoken agreement that the thing should exist. Nobody questions the premise. But I think it got worse with agents. You're sitting there with Cursor or whatever, building fast, and there's nobody pushing back.…