EVALS Workshop with Sam Alexander

​Hands-on workshop, limited to 20 seats

​AI is now a consumer of your design system. When it builds UI, it drifts away from your components and tokens, and your team pays for the rework. This workshop shows you how to measure that drift and reduce it.

​In a 1-hour session, you’ll run an eval against a public design system. You’ll see where AI-built screens use the system and where they ignore it. You’ll trace a score back to its cause, change one condition (the model, the prompt, or how the system is delivered to the AI), and run it again to see what changed. We track three scores: design fidelity, component adoption, and cost per quality outcome.

​Who should join?
Design system leads, design ops professionals, product managers, and anyone building or scaling design systems in enterprise environments.