Angelina YangFounder and Host, TwoSetAI · AI builder for visibility and discoverability

What is RAG?

3 Jul 2025 · 1 min

Watch on YouTube

0:51

A short explanatory transcript describing Retrieval Augmented Generation (RAG): embedding a user query, searching a vector database for relevant documents, and supplying those results to a large language model (LLM) together with a prompt so the LLM can generate an informed response. The speaker emphasizes RAG’s popularity and industry use and sketches a basic architecture flow.

Chapters

Rag AKA Retrieval Augmented Generation is a technique which is going to enhance LLMs by incorporating external knowledge bases and data sources before generating responses. This is a very popular and powerful technique used heavily in industry for different use cases. On the right hand side you can see a basic architecture of wrike. When the user query comes in, the first step is to embed the query and then search in the database vector database and find relevant information to this query from the vector database. Then bring that into the LLM along with the prompt and then LLM will generate hopefully a right answer for us. So that's how write is working.