Session
The Agent Worked. The System Didn’t.
It’s easy to assume that if an AI picks the right tool and generates the correct arguments, everything is fine. But what happens when a user says, "Cancel my order and refund me," the cancellation succeeds, but the refund times out?
What if a network drop causes the agent to retry or two requests try to cancel the same order simultaneously? What if your server crashes right in the middle of the workflow?
In this hands-on session, I'll be building a small AI-powered order operations agent and deliberately push it off the happy path. We’ll explore how retries, timeouts, partial failures and race conditions can turn a smart-looking agent interaction into a data integrity nightmare. Then, we’ll harden the system using idempotency, explicit state machines durable workflows and bulletproof compensation strategies.
Agenda:
1. The Illusion of Correctness
2. Breaking the Illusion
3. The Distributed-Systems Trap
4. Building the Guardrails
5. Live Demo
6. Designing Resilient Agentic Systems
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top