In this meet-up, we'll cover how to: Deploy agents with durable execution so runs survive crashes, deploys, and long waits for human input Use checkpoints and memory stores Add human-in-the-loop gates for consequential decisions We'll also discuss how teams