HLAB field notes

Retrieval Evaluation

Stories and analysis filed under Retrieval Evaluation.

01 stories
Terminal screenshot of the Claude RAG CLI answering a question and citing local source files

Benchmark retrieval quality

Turn six labeled retrieval questions into a repeatable BM25 quality report using Hit@k, Recall@k, and mean reciprocal rank, then use the report to find what lexical retrieval misses.