A basic introduction to Retrieval-Augmented Generation (RAG) with LangChain and Ollama, featuring a companion PDF and two hands-on notebooks.
| Section | Description |
|---|---|
| 1. 🔎 Retrieval-Augmented Generation (RAG) | Definition and overview Why it goes beyond a stand-alone LLM |
| 2. 💡 Concept | Core meaning of RAG Key benefits and features |
| 3. ⚙️ How to Build? | Guardrails Caching Monitoring Evaluation LLM Security |
| 4. 🏗 RAG System Design (Overview) | Input/Output orchestrator Retriever Data load & splitting Data conversion Storage component LLM setup Data indexing Prompt management |
| 5. 👀 Retriever: Indexing Pipeline | Data load & splitting Data conversion Storage (e.g., FAISS) |
| 6. 🧩 Generation Pipeline | Retriever: Query analysis & information retrieval Prompt Management: Contextual, few-shot, controlled, chain of thought LLM: Model configuration & generation flow |
| 7. 🔧 Hands-On |
Note: These two notebooks are the primary hands-on exercises for a basic RAG workflow. |
[1] LangChain: Retrieval. n.d. Available at https://python.langchain.com/docs/concepts/retrieval/
[2] Ollama: Library. n.d. Available at https://ollama.com/library
[3] Wolfe, Cameron R. The Basics of AI-Powered Vector Search. n.d. Available at https://cameronrwolfe.substack.com/p/the-basics-of-ai-powered-vector-search
This project is licensed under the MIT License.