Technology
Retrieval-Augmented Generation (RAG)
RAG is the pattern of giving an LLM access to your own data at runtime — fetching relevant context from a vector database, embedding it in the prompt, and letting the LLM generate an answer grounded in that data. Used for: customer support over your knowledge base, internal Q&A, document analysis.
More detail
Without RAG, LLMs answer from training data only — which is generic and stale. With RAG, you can ask 'what's our refund policy?' and get an answer grounded in your actual policy doc. Aiprosol's chat widget uses RAG against Supabase pgvector. Tools: pgvector (best for SMB), Pinecone (scale), Weaviate (hybrid search), LangChain/LlamaIndex (orchestration).