sahilsinha20

https://blog.lytix.co/posts/agentic_rag

Hey all! Thought this group might find this interesting - new approach to evaluating RAG pipelines using 'agents as a judge'. We got excited by the findings in this paper (https://arxiv.org/abs/2410.10934), about agents producing evaluations closer to human-evaluators, especially for multi-step workflows.Our first use case was RAG pipelines, specifically evaluating if your agent MISSED pulling...

https://blog.lytix.co/posts/ai-form-checker

Hey HN! Finally crossed off a nagging side project idea - using AI/ML to give realtime feedback on gym form (can be extended to any hobby where technique matters).I used posenet to capture coordinates in real-time, and GPT with some few-shot prompting to give feedback. Curious what ideas folks have for different/better approaches here!Looking for collaborators if anyone wants to hack on this...