AI and AgentsSimplified view
Multi-agent supervisor
A supervisor agent receives a request, hands sub-tasks to specialist agents one at a time, keeps shared memory, and combines their results into one answer.
- LangChain
- LangGraph
- Multi-agent systems
8 stepsPress play, or use next and previous to follow the flow one step at a time.
All steps
What this shows
In a supervisor design, one lead agent owns the conversation with the user and coordinates a team of specialists. The user talks to a single point of contact, while the specialists focus on narrow jobs.
Design notes
- Two memories, two jobs. What the user and lead agent have said is kept apart from the state the lead agent shares with the team.
- Sequential here, parallel when it pays off. The diagram delegates one agent at a time to keep it readable. Independent sub-tasks can run in parallel.
- A single voice. Because the supervisor combines the results, the user sees one coherent answer instead of three.
These patterns are covered in more depth in the book LangChain in Action. This is a simplified view.
More workflows
Test Automation
NL2Test, from a sentence to a Playwright test
How NL2Test turns a plain-language description into an executable Playwright test, with a model planning the test and an MCP server driving a real page.
10 steps 9 parts Simplified view
AI and Agents
How an MCP request flows
A walk through one Model Context Protocol request, from the client through authentication, memory and the model to a server that coordinates tools.
8 steps 8 parts Simplified view
AI and Agents
RAG pipeline
How retrieval-augmented generation works end to end, from loading documents into a vector store to answering a question with the closest passages.
8 steps 7 parts Simplified view