🎓 LangGraph: From Scratch - Divyansh Gangwar

citation: Source YouTube Video freeCodeCamp.org

Welcome to the LangGraph Deep Dive! This repo is designed to take you from "I've heard of AI Agents" to "I can build complex, reasoning, RAG-enabled systems" in 15 lectures.

The code here is written to be clean, readable, and heavily commented. It's built for students, by a (simulated) student. We start small and add complexity one variable at a time.

---

🚀 Getting Started

1. Prerequisites

You need:

• Python 3.10+ installed.
• A Google Gemini API Key (it's free!). Get one here.

2. Installation

Clone the repo and install dependencies.

# Clone the repo
git clone https://github.com/your-username/LangGraph-Course.git
cd LangGraph-Course

# Create a virtual environment (Recommended)
python -m venv .venv
source .venv/bin/activate  # Mac/Linux
# or
.venv\Scripts\activate     # Windows

# Install packages
pip install langgraph langchain langchain-google-genai python-dotenv chromadb pypdf

3. Setup API Keys

Create a .env file in the root directory and add your key:

GOOGLE_API_KEY="your_actual_api_key_here"

---

📚 Course Structure

Part 1: The Fundamentals (Notebooks)

We use Jupyter Notebooks for the basics so you can visualize the graph structure instantly.

• 1_HelloWorld.ipynb: Your first node. Input string -> Process -> Output string.
• 2_Exercise1.ipynb: Challenge! Build a personalized compliment bot.
• 3_MultipleInput.ipynb: Handling complex State (Lists, Ints, Objects).
• 4_Exercise2.ipynb: Challenge! Build a Calculator that chooses standard operations.
• 5_SeqGraph.ipynb: Building a pipeline (Node A -> Node B -> Node C).
• 6_Exercise3.ipynb: Challenge! The Resume Generator Pipeline.
• 7_ConditionalEdge.ipynb: The "Brain". Using routers to make decisions (If X, go to Node Y).
• 8_Exercise4.ipynb: Challenge! The Advanced Multi-Step Calculator.
• 9_LoopingGraph.ipynb: Building Cycles. Doing things repeatedly until a condition is met.
• 10_Exercise5.ipynb: Challenge! The "Guess the Number" Game (Cyclic Logic).

Part 2: Advanced Agents (Python Scripts)

We switch to standard .py files to build real-world agents with memory and tools.

• 11_AgentBot.py: Connecting a Real LLM (Gemini) to LangGraph.
• 12_AgentwMemory.py: Adding Memory so the AI remembers what you said.
• 13_ReActAgent.py: The ReAct Pattern. Teaching the AI to use Math Tools autonomously.
• 14_Drafter.py: Human-in-the-Loop. An agent that helps you write and save documents interactively.
• 15_RAGAgent.py: RAG. An agent that reads PDFs, remembers them (Vector Store), and answers questions.

---

🧠 Key Concepts You Will Learn

• StateGraph: The backbone of everything. It holds the "Memory" of your application.
• Nodes: Python functions that do work.
• Edges: The specific lines connecting nodes.
• Conditional Edges: Decision points (Routers).
• Cycles: Loops in the graph (vital for agentic behaviors).
• ToolNode: Giving the AI hands (calculators, search, file savers).
• Checkpointers (Memory): Persisting state across interactions.

---

🤝 Contributing

Found a typo? Want to add a cooler example? Open a Pull Request! We're all learning here.

Happy Coding! 🤖✨