OpenAI Agents SDK Handoffs | Deep Dive Tutorial
In OpenAI's Agents SDK, we can build multi-agent workflows in two ways. The first is agents-as-tools, which follow an orchestrator-subagent pattern. The second is using handoffs, which allow agents to pass control over to other agents. In this video, we'll dive into handoffs using Agents SDK and the latest LLMs, including gpt-4.1-mini and gpt-4.1 — using a multi-agent workflow containing a web search agent, dummy RAG agent, and code execution agent. 📖 Article: https://www.aurelio.ai/learn/agents-sdk-multi-agent 📌 Code: https://github.com/aurelio-labs/agents-sdk-course/blob/main/chapters/04-multi-agent.ipynb 🔗 LinkUp: https://app.linkup.so/?utm_source=james (Affiliate link) 👋🏼 AI Platform: https://platform.aurelio.ai 👾 Discord: https://discord.gg/c5QtDB9RAP Twitter: https://twitter.com/jamescalam LinkedIn: https://www.linkedin.com/in/jamescalam/ #ai #coding #aiagents #programming #artificialintellegence #python 00:00 Agents SDK Handoff 05:06 Code Start 07:12 Web Search Agent 10:14 RAG Agent 11:18 Code Execution Agent 11:46 Defining the Orchestrator 12:36 Agents SDK Handoffs 20:27 Using OpenAI Traces Dashboard 23:26 More Handoff Testing 26:20 Other Handoff Features 28:45 Agents SDK on_handoff 29:51 Agents SDK Handoff input_type 31:28 Agents SDK Handoff input_filter
