The challenges in using LLM-as-a-Judge - Sourabh Agrawal | Vector Space Talk #013

Qdrant's Vector Space Talks episode 13 with Sourabh Agrawal, CEO & co-founder of UpTrain AI, on LLM-based evaluation - why production-grade LLM-as-a-judge is much harder than prompting a model to grade a response, and how to use evals to improve RAG applications. Video: https://www.youtube.com/watch?v=vBJF2sy1Pyw