Defining Agent Quality for AI Engineers & PMs

​Most AI agent projects don't fail because the technology doesn't work. They fail because teams have no systematic way to define, measure, or prove quality before production finds the gaps.

​So it's time to fix that.

​In this webinar, Manouk Draisma (LangWatch CEO) and Rogerio Chaves (LangWatch CTO) will walk through a practical framework for defining agent quality from the ground up — before a single eval is written.

​By the end, you'll understand:

• ​Why shipping on intuition creates failures that are hard to reproduce and harder to fix

• ​How to specify agent behavior before writing code — what it should do, what it must never do, and how it should handle edge cases

• ​Who owns quality in practice and what the collaboration between engineers, domain experts, and product actually looks like

• ​How to build an evaluation framework using deterministic, model-based, and domain-specific grading

• ​How to set a launch threshold that gives your team confidence to ship

• ​How production failures should continuously sharpen your quality definition over time

​Whether you're building your first production agent, struggling with regressions after every model update, or trying to establish a quality baseline your team can actually trust — this session will give you a concrete process to move from intuition to systematic quality.