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
The secret ingredient for impactful AI products is “evals”—an architecture for ongoing evaluation of quality. Without evals, you don’t know if your output is good. You don’t know when you’re done. Because outputs are non-deterministic, it’s very hard to figure out if you are creating real value for your users, and when something goes wrong, it’s really tricky to figure out why. 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
"Activation isn’t just at the end of the journey; it must be embedded throughout the knowledge management process."
Maria TaylorKnowledge is Power: Managing the Lifeblood of the Design Org
October 3, 2023
"Democratization can be used arrogantly if we just let others do research so we can pick what’s fun for ourselves — Jared Spool."
Kathleen AsjesResearch Democratization: the Good, the Bad and the Ugly
March 10, 2022
"You have to build all the puzzle pieces with clear summaries before the patterns come together in mental models."
Indi YoungPaying Better Attention to the Problem with Indi Young
December 12, 2019
"If patients stop using the app because they’ve got their health under control, that’s like a dating app success — they got married, not that the app failed."
Daniel J. RosenbergDigital Medicine Design
September 26, 2019
"No one hated technologists, they just didn’t know who they were because there were no personal relationships cross teams."
Husani OakleyBias Towards Action: Building Teams that Build Work
June 14, 2018
"Our AI-supported mornings were no longer about just dotting the I’s and crossing the T’s, but deeper analysis sessions including clients."
Mujtaba HameedThe new horizon of ethnography: using AI to unlock the full potential of in-person research
March 11, 2026
"Think of AI not as a magic eight ball, but as a highly educated intern that can help with routine tasks."
Mark Interrante Harry MaxAI for Prioritization (3rd of 3 seminars)
July 11, 2024
"Our vision was to create an AI first tool for user research that delivers instantly well structured requirements and design inputs."
Joerg Beringer Thomas GeisScaling User Research with AI: Continuous Discovery of User Needs in Minutes
June 10, 2025
"I love getting other people in there. Shared ownership means people are going to reference it more often."
Gabrielle VerderberDocumentation Your Team Will Actually Use
October 3, 2023
