# 🎥 LangChain v0.3+ Crash Course: Build Real Agents, Deploy Production-Ready Apps Hi! 👋 This repo contains the code and project files from a YouTube playlist I’m making as I learn LangChain v0.3+ myself. The goal? To understand how the newer LangChain ecosystem works — and share what I’m learning through practical, step-by-step episodes. It’s aimed at developers who want to go beyond just playing with notebooks and start building real apps and agents using modern tools. --- ## 📌 What’s This Playlist About? LangChain has changed a lot recently, and many older tutorials are outdated or don’t reflect how things work now. This playlist is an attempt to: - Learn the **new LangChain Expression Language (LCEL)** and modular design - Build working projects using tools like **Poetry**, **LangGraph**, **LangServe**, and more - Explore things like deployment, caching, CI/CD, and RAG — not just toy examples I’m not an expert — I’m learning and figuring things out as I go. If you’re also trying to understand how to use LangChain properly in real dev setups, you might find this useful. Link to the PlayList : [LangChain v0.3+ Crash Course](https://www.youtube.com/playlist?list=PLv5fXmCxvRd_X3DHTl68Pra5_oigfpmhv) --- ## 📚 Episodes & Topics Covered | Ep# | Title | Focus | |-----|-------------------------------------------|-------------------------------------------------------------| | 1 | LangChain v0.3: What Changed & Why | What’s new, what broke, and why it matters | | 2 | Quickstart Project (Poetry Setup) | Set up a clean Python project using Poetry | | 3 | First LCEL Chain (Streaming + Async) | Prompt → model → parser with async/streaming | | 4 | LangGraph Basics | Build agent flows with conditional steps | | 5 | Deploy with LangServe | Serve your agent as an API with FastAPI | | 6 | LangSmith Integration | Add tracing and debugging | | 7 | RAG 2.0 (Retrieval-Augmented Generation) | Vector DBs, filters, and multi-doc retrieval | | 8 | Redis Caching for API Cost Control | Reduce OpenAI costs with caching | | 9 | Deploy on AWS Lambda | Serverless deployment with AWS | | 10 | CI/CD + Canary Releases | Add GitHub Actions and test deployment flows | | 11 | LangChain.js Parity (Bonus) | Try similar logic in JavaScript/TypeScript | | 12 | Capstone: Build & Deploy Real Agent | Wrap it all up into a working, deployed project | --- ## 🗂 Project Structure Each folder here matches one episode and includes the code or examples built during that part. Feel free to explore, run things, or use them as a starting point for your own learning. --- ## 🧰 Tools Used - Python 3.10+ - [Poetry](https://python-poetry.org/) for dependency management - LangChain v0.3+ - Optional: Redis, AWS CLI, Docker, etc. (used in later episodes) --- If you’re learning LangChain too, hope this helps. If you spot mistakes or have suggestions, feel free to open an issue or PR. I’m learning — and teaching — as I go. 🚀 # LangChainV0.3_CrashCourse