AI Agents and Multi-Agent Systems · Lesson 8
Introduction to LangGraph
LangGraph models an agent or workflow as a graph. Nodes are functions (call the model, run a tool), edges decide what runs next, and a typed state object flows through the graph.
Conditional edges allow branching and loops — for example, route back to the tool node while the model keeps requesting tools. Checkpointers save state so you can pause, resume, add human approval and keep conversation memory.
This explicit control makes agents easier to debug and more reliable than an opaque loop.
Key points
- Graph = nodes + edges + shared state
- Conditional edges enable loops and branches
- Checkpointers give persistence and human-in-the-loop
- Explicit control improves reliability