LangGraph - managing AI Agents: A Beginner's Guide

Get to grips with the LangGraph framework from theory to production-ready applications. Code examples are regularly updated to keep you abreast of the latest LangGraph and LangChain changes. Purchase of the print or Kindle book includes a free PDF eBook and all code samples. AI Agents can solve very complex situations and that’s why clear communication is essential to achieve the desired results. This is where LangGraph excels. Communicating with LLMs via code leads to far more reliable results than using only natural language (prompt engineering). AI Agents are LLMs on steroids. The anatomy of an agent consists of: AI_AGENT = LLM + MEMORY + TOOLS + PLANNING + DO_WHILE_LOOP The LangGraph framework is an excellent tool for implementing and orchestrating all of these components. Simply put, AI agents are LLMs with tools, that operate in a loop to accomplish specific goals. You can assign them tasks such as: a calendar manager to schedule and manage appointments, send reminders, and suggest optimal meeting times a personal AI Agent assistant to organize trips, book hotels and plane tickets for a destination in a given budget. These are much closer to the capabilities we expect from real AI. AI Agents can solve very complex situations thus clear communication is essential to achieve the desired results. This is where LangGraph excels. Communicating with LLMs via code leads to far more reliable results than using only natural language (prompt engineering). In this book, we’ll take you on a fun, hands-on journey where each chapter will focus on essential concepts like tool management or human-in-the-loop validation, while also coding practical implementations of these elements using LangGraph. What’s Inside AI Agents Introduction LangGraph Fundamentals Nodes, Edges, and Graphs in LangGraph Building Your First AI Agent Tool Calling and Functions AI Agents and Tools Using Stateful Graphs Threads in LangGraph ReACT Agents Human in the Loop Adding Interruptions Managing Permissions in the Graph State Multi Agent Systems Multimodal Agents (using video and audio) You can see this book as your launchpad. You'll build your first AI Agent within minutes, and slowly become competent and knowledgeable in this technology, with each chapter featuring a full practical example. Latest updates to the book: 24 Mar 2025: added 2 brand new chapters one on Multi Modal AI Agents (text and audio) and the other one on ReAct AI agents 15 Jan 2025: various typos and extra code samples 04 Dec 2024: added the Multi AI Agent Orchestration chapter Let's start learning!