Retrieval augmented generation, abbreviated RAG, is an AI architecture pattern that retrieves relevant documents from a knowledge source and includes them in the model's context window before generating an answer. The model's response is then grounded in the retrieved material rather than only its training data.
RAG is distinct from fine-tuning. Fine-tuning bakes knowledge into the model's weights through additional training. RAG keeps the model unchanged and supplies fresh, organization-specific knowledge at query time. Each has tradeoffs; for most organizational knowledge use cases, RAG is the more practical pattern.
The quality of a RAG system depends almost entirely on the quality of the underlying retrieval. A model with bad context produces bad answers, regardless of the model's capability.
Why Retrieval augmented generation (RAG) Matters for Distributed Teams
RAG is the dominant pattern for grounding AI in organizational knowledge. It is how most "AI that answers questions about your company" products work under the hood.
The teams that get RAG right invest heavily in the data layer — clean, structured, well-attributed source documents — because retrieval is only as good as what it retrieves.
Frequently Asked Questions
What is retrieval augmented generation?
Retrieval augmented generation, or RAG, is an AI pattern that retrieves relevant documents from a knowledge source and includes them in the model's context before generating an answer. It grounds responses in source data rather than relying only on model training.
Related Terms
Large language model (LLM)
A large language model, abbreviated LLM, is a neural network trained on massive amounts of text and code to generate and...
Read definitionAI hallucination
An AI hallucination is a confident, fluent, plausible-sounding answer from an AI system that is factually wrong. The sys...
Read definitionAI workflow
An AI workflow is a multi-step process that uses AI models for one or more of its steps. The workflow defines inputs, ou...
Read definitionGet the vocabulary that makes distributed teams work
One email per week on async governance. No spam.
See retrieval augmented generation (rag) in action.
StandIn is built around these concepts. Engineers publish declared state before going offline. The next shift starts with full context.