Tabish alam·13h agoEvaluating Retrieval in RAG SystemsA retrieval-augmented system can give a clear, fluent answer that is supported by the text it was shown, and still be wrong. The reason is…
Adnan Masood, PhD.·5d agoEvals All the Way Down: Fault Localization in Compound AI SystemsFlat LLM evals hide where pipelines break. Master RAG evaluation and RAG pipeline debugging across 5 layers from parsing to generation.
InAI Engineering CompassbyDipak Ahirav·Oct 25RAG Evaluation Explained: How to Know Whether Your AI Is Actually Retrieving and Answering…A RAG system can sound impressive while retrieving the wrong documents, citing irrelevant sources, or confidently generating unsupported…A response icon1
Sudhakar Tripathi·Oct 22Can You Trust an LLM Judge? A Hard Look at G-EvalWhat it gets right, where it quietly misleads, and how I use it with DeepEval
Adnan Masood, PhD.·Oct 21AI Metrics for Mere Mortals, Part 2: How to Measure Search, RAG, and Generated AnswersPrecision@K, Recall@K, Hit Rate@K, MRR, NDCG, Pass@K, RAG faithfulness, and answer correctness, each worked by hand on a small example
