Log in or create a free Rosenverse account to watch this video.
Log in Create free account100s of community videos are available to free members. Conference talks are generally available to Gold members.
Building impactful AI products for design and product leaders, Part 2: Evals are your moat
This video is featured in the AI and UX playlist.
Summary
Simply Put’s Peter van Dijck will demystify evals, and share a simple framework for planning for and building useful evals, from qualitative user research to automated evals using LLMs as a judge.
Key Insights
-
•
AI product development involves three layers: model capabilities, context management, and user experience, with evals central to experience quality assurance.
-
•
Automated evals help scale testing of AI with inherently open-ended inputs and outputs, enabling faster iteration cycles with confidence.
-
•
LLMs can serve as judges (evaluators) of other LLM outputs, which works because classification is cognitively easier than generation.
-
•
Defining what 'good' means for an AI system is a detailed, evolving process informed by research, domain expertise, and observed risks.
-
•
A three-option evaluation (e.g., yes/no/maybe) works better than fine-grained scales for consistent automated scoring by LLMs.
-
•
Synthetic data, generated by LLMs based on manually created examples, efficiently expands dataset breadth and usefulness.
-
•
Domain experts are essential for tagging data and establishing quality criteria, especially for high-stakes areas like healthcare or legal.
-
•
Building effective evals requires substantial effort—expect 20-40% of project resources devoted to this work.
-
•
Cultural differences impact subjective evals like politeness, requiring localization and careful domain definition.
-
•
AI product quality management is a strategic ongoing commitment, extending beyond initial development into production monitoring and iteration.
Notable Quotes
"AI products almost always have both open-ended inputs and outputs, which makes testing really hard."
"You have to build a detailed definition of what is good for my system to do meaningful automated evals."
"It’s much easier to classify an answer than to generate an answer, and that’s why LLM as a judge works."
"You don’t want to give too many options like rating from one to ten because consistency gets lost between different LLM calls."
"Synthetic data is useful because it’s easier to generate more examples of something you already have than to create entirely new data."
"If you launch in the US and politeness is an issue, first try to fix it with prompts; only if that fails should you build an eval."
"Evals are really your intellectual property—they define what good looks like in your domain."
"Domain experts are crucial for tagging data because users might say ‘that’s great,’ but experts can tell it’s totally wrong."
"You should plan 20 to 40 percent of your project budget on evals—it’s a lot more work than most people expect."
"This is where UX and product strategy bring huge value—defining what good means rather than leaving it to engineers alone."
Or choose a question:
More Videos
"By taking a moment to pause to just let people be people, we could allow the interview to resume with some dignity and purpose."
Susan Simon-DanielsWar Stories LIVE! Susan Simon-Daniels
March 30, 2020
"The cult of the new and shiny drowns out the quiet call of the well established."
Sharon BautistaTime to Make the Donuts: How User Research Helped Bridge Disparate Teams
January 8, 2024
"We’ve put much more effort into workshop facilitation and teaching how to use the space effectively than on the design system itself."
David Cronin Uday Gajendar Peter Morville Kendra ShimmellDiscussion
May 13, 2015
"We can no longer rely on interviews, surveys, and usability tests to answer the strategic, complex questions organizations face today."
Jemma Ahmed Megan Blocker Eduardo OrtizRedefining the research toolkit: Expanding methodologies for a changing world
March 12, 2025
"Designers connected with engineers closely, sometimes building templates themselves to quickly validate visions."
George Hinchliffe Joy LiuDelivering Amazing Experiences
June 10, 2021
"Many federal benefit programs face challenges verifying income, causing stress and leaving billions in unclaimed benefits."
Stefanie OwensOptimizing for Outcomes: Transformation Design in Systems at Scale
December 4, 2024
"I never record interviews with staff, never. Recording invites sugarcoating or lying because people are already vulnerable."
Steve Baty Richard Dalton Maria Giudice Harry MaxDiscussion
June 9, 2016
"Figma files are no longer just a collection of visual elements; designers actively shape product features now."
Ryan Matthew Alex KurchevBridging Design and Code: AI-Powered Design System Integration
September 11, 2025
"Conflict occurs when something important to us feels threatened or disrespected."
Laura WeissTurn Down the Heat: 3 Ways to Handle Conflict in the Moment
November 20, 2024
