Intro to Retrieval-Augmented Generation (RAG)

Retrieval-Augmented Generation (RAG) is a powerful way to build AI apps and Expert Models that are both privacy-aware and domain-specific. By running models locally and keeping your documents in your own environment, you can protect sensitive information, intellectual property, and internal knowledge—while still getting the benefits of advanced AI.

In this beginner-friendly workshop, we’ll show you how RAG works and walk through building a simple local app that can answer questions using your own files and data. Whether you’re working with product manuals, company knowledge bases, or research archives, this session will help you take the first step.

• Anatomy of a RAG system:

• Loading and splitting your documents

• Creating embeddings and storing them in a vector database

• Retrieving context and building AI-ready prompts

• Tools we’ll use:

• VS Code

• Ollama (for running language models locally)

• LangChain or LlamaIndex (to manage the RAG pipeline)

• Chroma or FAISS (for vector search)

• Live demo: Build a simple local RAG app using your own documents