RAG
Retrieval-augmented generation
Before answering, the system searches your documents and hands the relevant passages to the model. Answers get grounded in your data without retraining.
01In short
Before answering, the system searches your documents and hands the relevant passages to the model. Answers get grounded in your data without retraining.
In productionMost RAG problems are search problems: chunking, ranking, stale documents, permissions.
02Video
slot · videoRAG in 90 secondsadd: GUIDES['rag'].video
03Guide
slot · guide
A step-by-step guide for “RAG” goes here. Suggested outline:
- What it is — in one paragraph
- Why it matters in production
- How to do it — 3 to 7 steps
- Pitfalls we see in the field
04Checklist
slot · checklist
Four to eight things a team can tick before go-live.
05FAQ
slot · FAQ
The three questions clients actually ask about “RAG”.
06Related terms
Data & retrievalVector databaseA database built to store embeddings and find the nearest ones fast. The retrieval engine behind most RAG systems.Data & retrievalChunkingSplitting documents into pieces before embedding them. Too big and the answer drowns; too small and the context is lost.Data & retrievalRerankingA second pass that reorders search results by relevance before they reach the model.
Where we help · Build