Course
Applied LLM Systems: RAG, Agents & MCP
Build systems around a model: retrieve knowledge, call tools, and run agent loops.
L3 · AdvancedFast-movingKnown~3 h
What you’ll learn
- Explain why RAG grounds answers and reduces hallucination
- Describe embeddings, cosine similarity, and vector search
- Trace a full RAG pipeline from documents to assembled context
- Distinguish a single model call from an agent loop
- Explain tool use / function calling and the Model Context Protocol
- Tell a deterministic workflow apart from an agentic and an autonomous system
Prerequisites
llm-foundations
Module 1. Retrieval-Augmented Generation
Ground a model in retrieved documents: embeddings, vector search, and the full RAG pipeline.
Module 2. Agents and Tools
From a single model call to an agent that plans, calls tools, and reflects.
Module 3. MCP and Workflows
Connect models to tools and data with MCP, and place agents among deterministic workflows.