Hi and welcome to this course on building complex multi-agent teams and setups using LangGraph, LangChain, and LangSmith. In this course we’ll start from the ground up using LangChain, and then build and build, adding more complexity and tools as we go along. We will learn how to build a graph with paths, conditional paths, teams, team managers, and more, all stringing our agents together in powerful ways.
- In part 1, we’ll get started with the basics of LangChain, learning how to create prompt templates and Chains, working with the LangChain syntax to easily string together our LLM calls.
- In the next part we’ll learn how to write tools so that we can make our future agents powerful by giving them functions they can call. We will use the newest LangChain syntax for this and create both an image generation tool and a weather tool.
- Part 3 is where we will learn the basics of LangGraph, covering the underlying concepts and exactly how it works. We will learn by setting up our first agent and graph which can return a visual representation of the current weather in any city you name.
- In part 4 we’ll look at how we can take this all to yet the next level, discussing how we can use all of this to create a whole team of agents working together for us. We’ll also write a tool that can output PDF files in preparation for our multi-agent setup.
- Part 5 is where the rubber really hits the road and we will create a powerful multi-agent setup in LangGraph using a team, team manager, many agents, conditional paths, and more. We will create a team that can work together independently and create travel itineraries for us, providing them in PDF format with an inserted image and a full travel plan.
- In the final part we’ll have a look at writing asynchronous tools for our agents and then create a web research and article writing graph that can visit many web pages at the same time and then write an article about our desired topic for us.
I hope you’re as excited as I am to get started. Let’s dive in!