Glossary
RAG (retrieval-augmented generation)
A technique where the agent first retrieves relevant passages from your knowledge base and only then generates an answer based on them. It is what keeps the answer anchored in your content rather than in the model’s generic memory.
RAG exists to fix one problem: a language model asked about your business will answer anyway, from general knowledge, and it sounds exactly as confident when it is wrong. RAG inserts a retrieval step before generation, so the model writes from passages that came out of your own content.
In practice it is an indexing pipeline plus a search. Your sources get split into chunks, converted into vectors and stored; at question time the closest chunks are pulled and handed to the model together with the question.
Answer quality then depends less on the model than on the retrieval. If the right passage never comes back from the search, no model recovers it: which is why the state of your knowledge base matters more than which model sits behind it.
See also
Knowledge base
The set of sources the agent can consult: site, FAQ, product catalogue, documentation, spreadsheets. Answer quality is capped by the quality of this base, no agent invents what is not there.
Guardrail
A constraint applied to the agent to prevent answers that are out of scope, off-brand or ungrounded in your content. It covers faithfulness checks, topic blocking and masking of sensitive data.
Start the conversation today
Launch an AI agent that qualifies leads and books meetings around the clock. Live in under ten minutes.
A guided demo · your agent live in under 10 minutes