AI agents are moving fast from prototype to production, but scaling them reliably is a different game. In this session, you'll walk away with practical, hands-on experience across the full lifecycle of building AI agents that actually deliver business value.

We'll start with a production-ready agent template designed to scale from day one, so you're not reinventing the wheel every time you kick off a new project. From there, we'll dive into evaluation strategies that go beyond vibes, giving you a framework to measure whether your agents are doing what the business needs them to do and catching regressions before your users do.

We'll also cover how to tighten your inner dev loop. Local development workflows that let you iterate on agent behavior quickly without waiting on cloud round-trips or burning through API credits. Fast feedback, fewer surprises.

Finally, we'll put it all together by using AI assisted coding to build a brand-new agent live, showing how reusable skill definitions can accelerate agent development and keep your team shipping consistently.

Whether you're just getting started with AI agents or looking to level up what you've already built, this session gives you a concrete toolkit to build, evaluate, and iterate with confidence.

What you'll take away:

A production-ready agent template built for scale.
An evaluation framework to ensure your agents keep delivering real business value.
Local development workflows that speed up iteration cycles.
Deep dive experience using AI assisted IDEs to rapidly build new agents.