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.
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
"You owe it to your organization and your team to grow the practice while staying true to your vision."
Rob Mitzel Sébastien MaloThe Tale of Two Companies: Building a Successful UX Practice in a Century-Old Enterprise
January 8, 2024
"Thirty to fifty percent of total efforts are typically spent on rework, which saps morale and wastes resources."
Rima Campbell Amrit S BhachuIncrease Productivity and Drive Business Impact
September 24, 2024
"In 2019 and 2020, established design ops teams were incredibly rare; now they’re the number one response."
Bud CaddellTheme 2 Intro
September 30, 2021
"Leadership believing it’s about people and skills, opening up and unlocking talent—that was really beautiful."
Jaskiran KangWhy Community is Key to Professionalizing Design
October 28, 2022
"The DP&M tool has everything designers wish they could see from Jira and Figma, but often can’t because it’s either inappropriate or too hard to keep updated in those tools."
Ellie Krysl Jon FukudaPlanned Right. Managed Right. Designed Right.
June 6, 2023
"Aligning text with text and images with images sets an easy flow for scanning — mixing them breaks the user’s rhythm."
Billy CarlsonPro-level UI Tips for Beginners
September 9, 2022
"AI can serve as a conversational broker, helping non-designers interact naturally with design systems."
Louis Rosenfeld Billy Carlson Jon Fukuda Maria TaylorHow AI will Change DesignOps Tooling
October 3, 2023
"Always design a thing by considering it in this next largest temporal context — a day in a month, a decade in a century."
Sarah Auslander Betsy Ramaccia Gordon RossInsights Panel
November 18, 2022
"Accessibility is innovation and this statement could potentially be some unchecked ableism."
Alexis LucioScaling Accessibility Through Design Systems
June 9, 2022
Latest Books All books
Dig deeper with the Rosenbot
How are accessibility annotations in Figma integrated into AI-generated components?
What is the ethical complexity around deception in design, such as placebo buttons versus dark patterns?
What are effective first steps for starting research operations in a large organization with limited resources?