37 Things I Learned About Information Retrieval in Two Years at a Vector Database Company

A Weaviate engineer pulls back the curtain on two years of hard-earned lessons in vector search—breaking down BM25, embedding models, ANN algorithms, and RAG pipelines.

The real story? Retrieval workflows keep moving—from keyword-heavy (sparse) toward embedding-driven (dense). Across IR use cases, the industry’s letting go of crusty keyword matching.