Show HN: Active Source of Truth for Your Coding Agents

3 pointsposted 14 hours ago
by anphamthanh

2 Comments

gauravguitara

13 hours ago

what are your thoughts on using LLM as a judge for auto evals?

anphamthanh

13 hours ago

IMO, if you break down the criteria enough, LLM or not wouldn't be a problem. But if you feed a long answer to it and ask LLM if it's a good answer, then it will be hit or miss.