LangChain Master Class 🔗

A comprehensive LangChain tutorial repository with three real-world projects covering chains, agents, RAG, and ReAct patterns.

Projects

1. Ice Breaker 🧊

An app that takes a person's name and generates a summary and conversation starters using LangChain agents and LinkedIn data.

Key concepts: Agents, Tools, Output Parsers, LinkedIn API integration, Flask web UI

ice_breaker/
├── app.py                  # Flask web application
├── ice_breaker.py          # Main LangChain logic
├── output_parsers.py       # Custom output parsers
├── agents/
│   └── linkedin_lookup.py  # LinkedIn lookup agent
├── third_parties/
│   └── linkedin.py         # LinkedIn API client
├── tools/
│   └── tools.py            # Custom LangChain tools
└── templates/
    └── index.html          # Web UI

2. Medium Analyzer 📝

Ingests Medium blog posts and enables semantic search and Q&A over the content.

Key concepts: Document ingestion, text splitting, vector stores, retrieval chains

medium-analyzer/
├── ingestion.py        # Document ingestion pipeline
├── mediumblog1.txt     # Sample blog content
└── requirements.txt

3. ReAct LangChain 🧠

Implementation of the ReAct (Reasoning + Acting) pattern using LangChain agents with streaming callbacks.

Key concepts: ReAct agents, callbacks, streaming output, tool usage

react-langchain/
├── main.py             # ReAct agent implementation
├── callbacks.py        # Streaming callback handlers
└── requirements.txt

Tech Stack

Setup

git clone https://github.com/nofaukost/langchain-master-class.git
cd langchain-master-class

# Choose a project
cd ice_breaker  # or medium-analyzer or react-langchain
pip install -r requirements.txt

# Set environment variables
export OPENAI_API_KEY=your_key_here

License

MIT