Guides

From “what is a reranker?” to hybrid retrieval pipelines and evaluation — pick a starting point or read straight through.

Two-stage retrieval, why order matters, when reranking pays off.

Speed vs accuracy — and why production stacks use both.

Retrieve wide, rerank tight — with Python, LangChain, and latency tips.

Combine BM25 and vector search, then let the reranker fix the merge.

NDCG, MRR, labelled queries, and regression tests for your domain.

sentence-transformers, GPU sizing, and a minimal FastAPI service.

RAG vs support vs legal vs code — which model fits your use case.

When ColBERT beats bi-encoders and when you still need cross-encoder rerank.

Task instructions that change what “relevant” means — vs plain cross-encoders.

Seven reasons you see no lift — recall, pool size, truncation — and how to tell which one you have.

Wiring a rerank stage onto pgvector, Qdrant or Elasticsearch, with code.

Per-search or per-token? Which rerank API is cheaper at your volume.

Watch a cross-encoder reorder passages — bi-encoder proxy included.