Glossary
Retrieval-Augmented Generation (RAG)
A technique that retrieves relevant passages from your own documents and asks the model to answer using only those, so responses are grounded in your real content.
A raw language model answers from its training data, not your business. RAG fixes that by first retrieving the most relevant chunks from your own documents, then giving them to the model as the source of truth for its answer.
Done well, RAG lets an assistant cite your real content and say 'I don't know' when the answer isn't there, which is exactly the behaviour that makes a chatbot trustworthy. Done badly, it retrieves the wrong context and the model confidently answers from it.
How we use it
RAG is the backbone of the grounded chatbots and internal assistants we build, paired with evaluation so we can measure whether the answers are actually correct.
Related terms
Get in touch
Want to put this into practice?
If this concept is relevant to something you're building, a short note is the fastest way to get practical help.