Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn how to build multi-agent systems using OpenAI's Agents SDK in this comprehensive tutorial. Explore the agents-as-tools method to create an orchestrator-subagent system while utilizing OpenAI's gpt-4.1 and gpt-4.1-mini models. Follow along to develop a complete multi-agent workflow capable of searching the web, utilizing internal documents, and executing code. The tutorial covers Python setup, creating various subagents (orchestrator, web search, RAG, and code execution), implementing the main orchestrator agent, evaluating the workflow's performance, and discussing the advantages and limitations of orchestrator-based approaches. Access the accompanying code repository, article, and community resources to enhance your learning experience.
Syllabus
00:00 OpenAI's Agents SDK
01:38 Python Setup
02:51 Orchestrator Subagent
05:57 Web Search Subagent
11:40 RAG Subagent
17:15 Code Execution Subagent
23:44 Orchestrator Agent
28:44 Evaluating our Multi-Agent Workflow
39:31 Pros and Cons of Orchestrators
Taught by
James Briggs