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.
Hands-on AI #1: Let’s write your first AI eval
This video is featured in the Evals + Claude playlist.
Summary
If you’re a product manager, UX researcher, or any kind of designer involved in creating an AI product or feature, you need to understand evals. And a great way to learn is with a hands-on example. In this talk, Peter Van Dijck of the helpful intelligence company will walk you through writing your first eval. You will learn the basic concepts and the tools, and write an eval together. This talk is hands on; you can follow along, and there will be plenty of time for questions. You will go away with an understanding of the basic building blocks of AI evals, and with the confidence that you know how to write one. And more importantly, you’ll build some intuition, some product sense, around how the best AI products today are built, and how that can help you use them more effectively yourself.
Key Insights
-
•
Evals consist of a task, a golden dataset with known correct outputs, and an evaluator that measures correctness.
-
•
Manual AI prompt testing is slow and inconsistent; automated evals accelerate and scale evaluation.
-
•
UX and product teams can and should learn evals as a practical, non-technical skill.
-
•
Creating your own golden dataset is essential and cannot be outsourced or fully automated.
-
•
Models are fixed once trained; improvements happen by refining prompts and context design, not retraining the model.
-
•
Evaluations measure task performance, not the underlying model itself, allowing comparison across models.
-
•
Outputting a confidence score from models is unreliable due to lack of internal memory and inconsistent scale interpretation.
-
•
Biases are baked into models during training via evals used in post-training refinement.
-
•
LLMs can be used to judge other LLM outputs to evaluate tasks with non-binary answers.
-
•
Effective eval work requires collaboration across data analysts, engineers, subject matter experts, and UX/product teams.
Notable Quotes
"Evals are like a way to define what good looks like."
"The model was baked and once it’s baked, it does not learn again until they bake a new one."
"You need to be looking at the data. Nobody wants to, but that’s core work."
"Without a golden dataset, you have to build the golden dataset yourself."
"We’re not teaching the model anything; we’re improving our prompts and context."
"Confidence scores from the model are not a good idea because the model has no memory."
"Biases are baked in through the evals used during model training and post-training."
"LLMs judging other LLMs might sound crazy, but if you do it right, it works."
"Evals are a product and UX skill; learning them lets you make these systems do what you want."
"There is a large and growing capability overhang in these models we haven’t discovered yet."
Or choose a question:
More Videos
"Try to make your component hierarchy as flat as possible because tall, nested structures become hard to navigate."
Jaime CreixemsBest Practices when Creating and Maintaining a Design System
June 7, 2023
"Finding the balance between transformation and preservation is a delicate issue in Istanbul."
Onur Kocan Ayhan EnsiciUnderstanding the Strategy for Civic Design in a Complex City: Istanbul
November 16, 2022
"Many teachers told us more than half their middle schoolers have first-hand experience with guns."
JD BuckleyCommunicating the ROI of UX within a large enterprise and out on the streets
June 14, 2018
"Design for one extent to many is Microsoft’s inclusive design motto that guides effective accessibility."
Amber KnablEmpowering innovation: The critical role of inclusive product development in the AI era
June 4, 2024
"Working with AI is a back and forth negotiation, not a vending machine where one prompt gets you the answer."
Kritika SonyMoving AI offscreen: Exploring failures, constraints, and recovery in physical game design
June 10, 2026
"Always design your high-impact slides for the busy executive in the back of a limo."
Mujtaba HameedFrameworks for Excellence: Using Visual Thinking and Communication to Elevate Your Research
March 26, 2024
"More features and functions do not equal a better product."
Catherine CourageThe Enterprise UX Journey: Lessons From the Voyage & The Opportunity Ahead
May 13, 2015
"What we practice at a small scale is a pattern for the whole system—fractal problems repeat across teams and orgs."
Dominique WardThe Most Exciting Time for DesignOps is Now
September 8, 2022
"100% of people responded to our care kits offer and wrote back about their experience. That sparked community connection during isolation."
Michelle MorrisonCulture Design
May 21, 2020
