Build a Multi-Agent Research Team · 30 min · Python · LangGraph

Multi-Agent Architecture — Why Graphs, Not Chains

A chain runs step 1, then step 2, then step 3. A graph can run steps 2, 3, and 4 at the same time and wait for all of them -- and that difference is the entire reason this course needs 5 agents instead of 1 long prompt.

Hiring signal: Multi-agent orchestration with LangGraph is one of the dominant production patterns in serious AI deployments -- knowing the graph model at the level of nodes, edges, and shared state (not just 'I called LangGraph once') is what separates people who've used it from people who understand it.

What you will learn

Explain why a graph model supports parallelism and conditional routing that a linear chain can't
Draw your own agent graph (nodes = agents, edges = data flow) before writing code
Build and run a trivial 2-node LangGraph graph and inspect the resulting state

Introduction

By the end of this course, a research question you actually care about goes in, 5 specialized agents work on it — 3 of them simultaneously — and a cited Markdown report comes out in 2-4 minutes. Today's question: why does this need to be 5 agents in a graph, instead of one long, well-crafted prompt?

What You're Building

A trivial 2-node LangGraph graph — no research logic yet. The point is seeing the mechanics (nodes, edges, state) work on something simple enough to fully understand before Lesson 2 makes the state schema real.

Unlock the full lesson

You've read the first 2 sections. The rest of this lesson covers Chains run in order. Graphs don't have to., The trivial graph, What you're building today — plus a hands-on lab, quiz, and project artifact.

Create a free account to unlock Phase 0 and Phase 1 of every course — no credit card.

Key terms in this lesson

Agent