Staging environment
Free Lesson

Designing AI Evals for UX & Product Teams

Hosted by Melissa Wittmayer

Wed, Sep 23, 2026

7:00 PM UTC (30 minutes)

Virtual (Zoom)

Free to join

306 students

Invite your network

What you'll learn

Map the experience end-to-end

Map how the workflow, agent, the systems it interacts with, and the humans involved work together to accomplish a goal.

Identify and prioritize the top risks

Surface where and how the experience could fail before you start building evaluations.

Define the right focus for your evaluations

Translate the experience you want to create into specific evaluation dimensions and observable behaviors.

Why this topic matters

Agentic AI has changed what designers and product builders need to own. Evals are critical to building reliable agents, but teams often benchmark agents against how humans perform a task without first understanding what makes the experience successful. This hands-on course gives UX and product teams a practical framework for mapping the experience, identifying risks, and defining what to evaluate and why before building their evals.

You'll learn from

Melissa Wittmayer

Design Leader for agentic UX & evals, speaker, writer

Melissa Wittmayer is a Director of User Experience in AI with over 14 years of experience in human-centered design across higher education, non-profit, and technology. She specializes in human-AI collaboration, agentic UX, and designing evaluation frameworks that help teams measure what matters to users. With a background in human services and crisis intervention, she brings a human-centered approach grounded in safety, trust, and meaningful experiences.

See all products from Melissa

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.