This repository provides a hands-on course for developing LLM-powered AI agents using LangGraph. It targets developers aiming to build production-grade AI agents, offering practical examples of RAG, ReAct agents, reflection, and multi-step graph architectures. The benefit is accelerated learning and development of complex AI agent systems.

How It Works

The course structure is built around Git branches and commits, where each branch represents a distinct project and each commit signifies a learning step. This granular approach allows users to follow along with the development process commit-by-commit, fostering a deep understanding of LangGraph's capabilities for building agentic workflows.

Quick Start & Requirements

• Install: Clone the repository, checkout a project branch (e.g., git checkout project/agentic-rag), install dependencies with poetry install, and run with poetry run python main.py.
• Prerequisites: OpenAI API key is mandatory. Tavily API key and LangChain API key (with LANGCHAIN_TRACING_V2=true) are optional for web-search and tracing respectively.
• Setup: Requires cloning the repository and setting up API keys.
• Links: Repository Map, Lesson-By-Lesson

Highlighted Details

• Each branch is a project, and each commit is a lesson, enabling granular learning.
• Covers advanced RAG with self-correction and adaptive routing.
• Implements ReAct (Reasoning + Acting) agents and Reflection/Reflexion agents.
• Demonstrates multi-step graphs with conditionals, parallelism, and web-search tools.

Maintenance & Community

• Active community support via Discord for Q&A and project feedback.
• GitHub Issues are welcomed for bugs and feature ideas.
• Acknowledgement of the LangChain/LangGraph team.

Licensing & Compatibility

• The repository itself is not explicitly licensed in the README.
• Compatibility with commercial use or closed-source linking is not specified.

Limitations & Caveats

The repository's licensing is not clearly defined, which may impact commercial adoption. Specific hardware or software version requirements beyond API keys are not detailed.