Build production-grade AI agents—fast. This repository is the hands-on companion to my Udemy bestseller. Every branch is a project, every commit is a lesson. Clone it, code along, and ship your own LangGraph agents.
- Agentic RAG – Retrieval-Augmented Generation with self-correction & adaptive routing
- ReAct Agent – Reasoning + Acting loop implemented in LangGraph
- Reflection & Reflexion Agents – Agents that critique and improve themselves
- Multi-Step Graphs – Complex flows with conditionals, parallelism, and web-search tools
| Branch | Project Snapshot | Live Code |
|---|---|---|
| project/agentic-rag | Advanced RAG pipeline with grading, web-search & adaptive routing | link |
| project/ReAct-agent | Classic ReAct (Reason + Act) agent in LangGraph | link |
| project/reflection | Minimal reflection demo – the why before the wow | link |
| project/reflection-agent | Full reflection agent that revises its own output | link |
| project/reflexion-agent | Reflexion agent that learns from past runs | link |
✨ Tip: Checkout a branch, then use
git log --onelineto watch the lessons unfold commit-by-commit.
| # | Commit | Lesson Title | Key Skill |
|---|---|---|---|
| 1 | 5b2b18e |
Project Kick-Off | Repo & env setup |
| 2 | 2693185 |
Folder Structure 101 | Clean project scaffolding |
| 3 | 513e3cf |
Ingestion Pipeline | Load & embed data |
| 4 | 03f79ae |
Graph State | Passing memory between nodes |
| 5 | c2d71c7 |
Retrieve Node | Context fetching with LangGraph |
| 6 | 9107e7a |
Grade Docs Node | Structured relevance filtering |
| 7 | 6d4fdc4 |
Web Search Node | Tavily API integration |
| 8 | bc57b63 |
Generation Node | Prompting & LLM calls |
| 9 | a450f9b |
Wiring the Graph | Fan-in, fan-out, conditionals |
| 10 | 5400fb7 |
Self-RAG | Let the LLM critique itself |
| 11 | 034e53f |
Adaptive Router | Dynamic tool selection |
Feel free to cherry-pick commits or rewind with git checkout <hash> to experiment.
# 1. Clone & enter
$ git clone https://github.com/emarco177/langgaph-course.git
$ cd langgaph-course
# 2. Choose a project branch
$ git checkout project/agentic-rag # for example
# 3. Install deps (Poetry)
$ poetry install
# 4. Run
$ poetry run python main.pyCreate a .env file:
OPENAI_API_KEY=...
TAVILY_API_KEY=... # optional – for web-search lessons
LANGCHAIN_API_KEY=... # optional – for LangSmith tracing
LANGCHAIN_TRACING_V2=true # optional
PYTHONPATH=$(pwd)Looking for your next role building Agents, RAG apps, or LangGraph integrations? Check the LangJobs.dev board—built for the 150 000-strong AI-Engineering community.
Post a job or explore openings that use the exact tech stack you're mastering here.
- 💬 Discord – Private server for Q&A and project feedback
- 🐙 GitHub Issues – Bugs, ideas, PRs welcome!
- 🛠️ Bonus Repos – Extra templates & utilities for course alumni
All future updates are free. Your one-time enrollment grows with the ecosystem.
Enjoy the course & happy coding! 🎉
Big thanks to the LangChain / LangGraph team and their excellent documentation and tutorials that make this course possible.