RAG & MCP Fundamentals – A Hands-On Crash Course

This practical crash course teaches you to build integrated AI systems rather than standalone tools. You will first master Retrieval-Augmented Generation (RAG) to connect models to your own data for accurate, context-aware answers. Next, you will learn the Model Context Protocol (MCP) to coordinate communication and actions across multiple software components. By the end, you will know how to use RAG for knowledge and MCP for system-level coordination to create sophisticated, multi-part applications. Hands-on Labs: https://kode.wiki/4nzgBbW ❤️ Support for this channel comes from our friends at Scrimba – the coding platform that's reinvented interactive learning: https://scrimba.com/freecodecamp ⭐️ Contents ⭐️ Part 1: Retrieval Augmented Generation (RAG) - 0:00:00 Course Overview: Building Integrated AI Systems - 0:00:52 The Simplest Explanation of RAG - 0:01:50 Real-world Use Case: Internal Policy Chatbot - 0:02:41 Understanding Retrieval, Augmenting, and Generation - 0:04:12 When to use Prompt Engineering, Fine-Tuning, or RAG - 0:06:40 Solving Voice and Style with Fine-Tuning - 0:08:15 Why RAG is Best for Dynamic Factual Information - 0:09:36 Keyword Search Techniques: TF-IDF and BM25 - 0:13:12 Hands-on Lab 1: Basic Search and Keyword Limitations - 0:16:10 Introduction to Semantic Search and Meaning - 0:18:01 Embedding Models: Parameter Size and Local vs. API Models - 0:20:04 How Embeddings Convert Text into Mathematical Vectors - 0:22:00 Vector Similarity and the Dot Product - 0:27:08 Hands-on Lab 2: Implementing Semantic Search with Embedding Models - 0:30:54 Scaling with Vector Databases: Chroma and Pinecone - 0:32:15 Indexing Algorithms: HNSW, IVF, and LSH - 0:35:53 Hands-on Lab 3: Initializing and Querying a Vector Database - 0:39:21 The Precision Problem: Why Document Chunking is Essential - 0:41:40 Chunking Strategies: Fixed-size, Overlap, and Boundary Rules - 0:44:28 Hands-on Lab 4: Document Chunking and Optimized Retrieval - 0:48:49 Bringing it All Together: The RAG Pipeline - 0:50:40 Hands-on Lab 5: Building a Complete End-to-End RAG Pipeline - 0:52:54 Production Concerns: Caching, Monitoring, and Error Handling - 0:54:00 Implementation Strategies for Query, Embedding, and LLM Caching - 0:56:51 Essential Metrics for Tracking RAG Performance - 0:57:39 Production Architecture: Microservices on Kubernetes Part 2: Model Context Protocol (MCP) - 0:59:40 Introduction to Model Context Protocol (MCP) - 1:00:59 The Role of AI Agents in Action-Oriented Systems - 1:06:15 Why We Need Standardized Tools for Third-Party Interactions - 1:08:00 MCP Architecture: Clients, Servers, and Local vs. Remote Hosting - 1:10:12 Hands-on Lab 6: Setting up the AI Assistant Environment - 1:12:40 Core MCP Components: Resources, Tools, and Prompts - 1:14:13 Understanding the MCP Specification and JSON-RPC Protocol - 1:22:55 Hands-on Lab 7: Connecting to and Using an Existing MCP Server - 1:26:22 Building a Custom MCP Server with the Python SDK - 1:29:12 Testing with the MCP Inspector - 1:29:45 Hands-on Lab 8: Developing Resources, Tools, and Prompts for MCP - 1:35:29 Building an MCP Client: Roots, Sampling, and Elicitation 🎉 Thanks to our Champion and Sponsor supporters: 👾 @omerhattapoglu1158 👾 @goddardtan 👾 @akihayashi6629 👾 @kikilogsin 👾 @anthonycampbell2148 👾 @tobymiller7790 👾 @rajibdassharma497 👾 @CloudVirtualizationEnthusiast 👾 @adilsoncarlosvianacarlos 👾 @martinmacchia1564 👾 @ulisesmoralez4160 👾 @_Oscar_ 👾 @jedi-or-sith2728 👾 @justinhual1290 -- Learn to code for free and get a developer job: https://www.freecodecamp.org Read hundreds of articles on programming: https://freecodecamp.org/news