1Most teams optimize the prompt. Agentic systems have more moving parts (opens in new tab)(aevyra.ai)3agunapal1mo ago0Save
2Show HN: Verdict – model evals on your own data, not someone else's benchmark (opens in new tab)(github.com)GitHub2agunapal1mo ago0Save