News for hackers. Stuff that matters.
Re:Claude Sonnet 5 by sanderjd on Tuesday June 30, 2026 @04:05PM (#48738487)
Maybe. But that sounds like a large amount of bespoke work for what seems like a common problem?
Re:Claude Sonnet 5 by manojlds on Tuesday June 30, 2026 @04:41PM (#48738885)
I was talking about enterprise agents and then realized the question is more about coding agents.
Re:Claude Sonnet 5 by sanderjd on Tuesday June 30, 2026 @04:53PM (#48739049)
Ah I see! Yes, I was talking about a coding harness, not an enterprise agent. I entirely agree with you that your suggestion of driving it via evals is the right thing for that use case!