A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluationarxiv.org4 points·by jflynt76·8 gün önce·0 comments
Multi-Agent Simulation Framework for Verifiable Synthetic Corporate Corporaarxiv.org4 points·by jflynt76·11 gün önce·0 comments
Why most AI evals would miss the Linear sales email failuretenureai.dev6 points·by jflynt76·19 gün önce·2 comments
Two AI judges scored our agent's answer 0.85, but it never opened the filetenureai.dev6 points·by jflynt76·20 gün önce·0 comments
Show HN: Synthetic corporate dataset generator for AI agent evaluationgithub.com3 points·by jflynt76·30 gün önce·0 comments
Show HN: Tenure – Traceable AI memory with configurable memory modestenureai.dev3 points·by jflynt76·geçen ay·0 comments
You Don't Need a GitHub Copilot Subscription to Use VS Code AI Featuresmedium.com5 points·by jflynt76·geçen ay·1 comments
AI Memory Proves Inefficient: Tenure Project Detects 95% Error Ratezamin.uz5 points·by jflynt76·geçen ay·0 comments
jflynt76·19 gün önce·discussFor what it's worth, I didn't describe it as anything; just posted the link. It's a paper with open code, no product behind it.
jflynt76·geçen ay·discussI think it's because too many people have released tools that's clearly not ready for production because they don't know what to actually check. So it's now just easier to pattern match away any good tools that might surface.