Build Your First AI Agent
Agents, RAG, and shipping to production - the builder's path.
Free to watch. Starting adds a personal copy to My Learning with progress tracking and a curated-path certificate on completion.
About this path
Go from agent theory to a deployed system: agent fundamentals and design patterns, multi-agent protocols (MCP, A2A), retrieval-augmented generation from scratch, and the deployment lessons (FastAPI, long-term memory) that turn a demo into a product. Ends with a capstone build.
What you'll learn
- Design agents with tool use, planning, and memory
- Explain and apply agentic protocols like MCP and A2A
- Build a RAG pipeline from scratch and know why RAG fails
- Deploy an agent behind an API with long-term memory
- Ship a capstone research-assistant agent
The path, module by module
Each step is lifted from a full course in our catalog, and where a course repeats itself we keep only the lessons that earn their place.
Module 1 · from AI Agents for Beginners
What an Agent Is, and How to Design One
Microsoft's four core lessons on agents, tool use, and context engineering. Their playlist also carries a 64-minute re-upload of all ten; you skip it.
4 lessons · 34m · hand-picked from 9
What are AI agents? · How to design good AI agents · What Is the Agent Tool Use Design Pattern?
Module 2 · from AI Agents for Beginners
Protocols, Planning, and Memory
Agentic RAG, MCP and A2A, multi-agent systems, planning, and memory. The five that separate an agent from a chatbot with tools.
5 lessons · 48m · hand-picked from 9
What Is agentic RAG? · Using Agentic Protocols (MCP, A2A, and NLWeb) · How to use a multi-AI agent system
Module 3 · from RAG From Scratch
RAG, Part by Part
Indexing, retrieval, generation, and query translation, taught end to end by LangChain themselves.
8 lessons · 47m
RAG From Scratch: Part 1 (Overview) · RAG From Scratch: Part 2 (Indexing) · RAG From Scratch: Part 3 (Retrieval)
Module 4 · from LangChain & LangGraph: RAG to Multi-Agent
LangChain in Practice
LCEL, a vector store, and a knowledge-base chatbot. Enough to build, without the four-hour production courses beside them.
3 lessons · 1h 16m · hand-picked from 8
LangChain Expression Language (LCEL) | Langchain Tutorial | Code · Using Langchain and Open Source Vector DB Chroma for Semantic Search with OpenAI's LLM | Code · Chatbot Answering from Your Own Knowledge Base: Langchain, ChatGPT, Pinecone, and Streamlit: | Code
Module 5 · from LangChain & LangGraph: RAG to Multi-Agent
Ship a LangGraph Agent
A LangGraph agent, then the FastAPI production setup with sessions and history. This is the step tutorials usually leave out.
2 lessons · 1h 20m · hand-picked from 8
LangGraph RAG Agent Tutorial | Basics to Advanced Multi-Agent AI Chatbot | With Code · Integrating LangGraph RAG Agent with FastAPI | Production Setup with Sessions, History, Vector DB
Module 6 · from AI Agents for Beginners
Capstone Project
Build a research assistant agent. A capstone that forces every earlier concept into one artifact.
1 lessons · 2m
Build: A Research Assistant Agent