Loading video player...
AI agents call LLMs, execute tools, and often run for long periods of time. But when something fails mid-run, you typically lose everything: the conversation history, the progress, and any expensive operations already completed. In this video, we show how the OpenAI Agents SDK integration with Temporal brings durability and resilience to your agents. First, we interrupt a standard agent and watch all state disappear. Then we run the same scenario with the Temporal integration: - The agent process stops - The Temporal Workflow continues running - Every completed step and result is still preserved - Restarting the agent brings it back exactly where it left off - No reruns, no duplicated work, no lost progress The best part: you write normal OpenAI Agents SDK code. Temporal automatically adds durability behind the scenes. Your agent can stop, your network can cut out, and your workflow continues safely. If you’re building production-grade agents, check out the OpenAI Agents SDK + Temporal integration and the full walkthrough in this video: https://youtu.be/fFBZqzT4DD8 --- Temporal is the simple, scalable, open source way to write and run reliable cloud applications. Learn more Blog: https://temporal.io/blog How Temporal Works: https://temporal.io/how-temporal-works Community Slack: https://temporal.io/slack Developer resources Docs: https://docs.temporal.io Courses: https://learn.temporal.io/courses Support forum: https://community.temporal.io