Getting Started with LangGraph
Published by O'Reilly Media, Inc.
Hands-on introduction to LangGraph for building multi-agent workflows
Course outcomes
- Understand the core features of LangGraph
- Learn how to integrate LangGraph with the LangChain ecosystem
- Develop skills to build multi-agent applications using LangGraph
- Manage state across multiple interactions and dynamic routing in workflows
- Explore practical applications and advanced workflows powered by LangGraph, such as multi-agent research workflows, scheduler agents, and multi-agent coding workflows
Course description
Join expert Lucas Soares for an in-depth, practical introduction to LangGraph, a framework designed to work in conjunction with LangChain for building multi-agent workflows powered by large language models like ChatGPT. You’ll learn how to create complex, multi-agent applications that leverage LangGraph and the LangChain ecosystem and how to implement state management and dynamic routing within LangGraph workflows. You’ll also get a hands-on introduction to building applications such as scheduler and coder agents.
What you’ll learn and how you can apply it
- Understand the absolute basics of LangChain the core features of LangGraph
- Learn how LangGraph integrates with LangChain seamlessly
- Create multi-agent applications using LangGraph
- Implement state management and dynamic routing within LangGraph workflows
- Explore practical applications and case studies involving LangGraph
This live event is for you because...
- You’re a software engineer or developer
- You’re a data scientist or machine learning engineer
- You’re interested in learning about large language models, specifically ChatGPT, and how to build applications.
Prerequisites
- Familiarity with Python programming
- A computer with Python installed locally or set up on a Google Colab notebook
- Some knowledge of machine learning concepts
- Basic understanding of LLMs
- Basic understanding of LangChain
Recommended preparation:
- Review the absolute basics of LangChain and agents
- Take Getting Started with LLM Agents Using LangChain (live online course with Lucas Soares)
- Take Getting Started with LangChain (live online course with Lucas Soares)
Recommended follow-up:
- Explore advanced features of LangGraph and LangChain
- Take Getting Started with Llama3 (live online course with Lucas Soares)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Introduction to LangGraph (30 minutes)
- Presentation: Introduction to LangChain absolute basics; introduction to LangGraph—from persistence to configuration
- Hands-on exercise: Build an agent workflow
- Q&A
- Break
Integrating design patterns in LangGraph (30 minutes)
- Presentation: Subgraphs, branching, and human-in-the-loop; tool calling, dynamism, and structured formats
- Hands-on exercise: Create an agentic research workflow
- Q&A
- Break
LangGraph Studio for Debugging and Visualization (40 minutes)
- Presentation: Demo of Real-time visualization, monitoring, and debugging in LangGraph Studio
- Hands-on Exercise: Observing multi-agent interactions in LangGraph Studio
- Q&A and Break
Agentic Rag with LangGraph (40 minutes)
- Presentation: Collaboration, supervision and hierarchy; RAG concepts with LangGraph
- Hands-on exercise: Build a coding agent; build an RAG agent over webpage
- Q&A
- Break
Building Local Agentic Workflows with LangGraph (50 minutes)
- Presentation: Building a web navigation agent; building agents that can schedule and execute tasks autonomously
- Hands-on exercise: Build a planner agent that can integrate with or without human-in-the-loop
- Q&A
LangGraph Deployment (50 minutes)
- Presentation: Deploying a LangGraph Application
- Hands-on exercise: Best practices and template code for deploying langgraph applications
- Q&A
Your Instructor
Lucas Soares
Lucas Soares is a machine learning engineer who has worked at K1 Digital and Biometrid, where he developed computer vision and NLP models for applications such as document verification, OCR-based applications, and recommender systems. Lucas has also developed various ML models, including neural networks, Siamese networks, convolutional neural networks, LSTMs, and genetic algorithms.
Skills covered
- Large Language Models (LLMs)
- Graph Analytics