In spite of all the recent panic, nothing has really changed about software development. We’re worried about agents being unreliable. Newsflash: human devs are too. We’re worried about agents producing more code than we can read, but did you ever really do all your reviews within 24 hours? We’re worried about the quality of agentic code, but you also had that one co-worker…

Pull requests, code reviews, tickets, and 1:1s are the "harness" human devs have built to trust each other's work, and AI-forward engineering teams are building all these processes into coding harnesses for agents. But trust alone can't ensure correctness.

That makes verification the next real unlock for agentic coding workflows. Effective verification provides fast, trustworthy, and legible correctness feedback that drives agents and human developers alike to converge on the spec with fewer blind alleys and hidden bugs. We'll show how you can use deterministic simulation testing – whole-system property-based testing in a fully reproducible environment – to ship better, more ambitious changes, faster.