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 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
"We couldn’t even begin to demonstrate our team’s impact unless we could answer the question compared to what."
JD BuckleyCommunicating the ROI of UX within a large enterprise and out on the streets
June 14, 2018
"Supporting designers with basic tools like a working laptop can be surprisingly complex but hugely impacts efficiency."
Jose Coronado Julie Gitlin Lawrence LipkinPeople First - Design at JP Morgan
June 10, 2021
"Systems thinking requires constant engagement with stakeholders; working in isolation will not lead to successful outcomes."
Sheryl CababaExpanding Your Design Lens with Systems Thinking
February 23, 2023
"If your priorities are roughly aligned with organizational priorities, you’ll be making good trade-offs."
Harry Max Jim MeyerPrioritization for Leaders (2nd of 3 seminars)
June 27, 2024
"If we do X, then Y percent of people will do Z—that’s how you frame hypotheses to learn from both success and failure."
Alissa BriggsHow to Coach Enterprise Experimentation
May 14, 2015
"Starting with positives and acknowledging good work helps break down resistance when sharing findings that reveal weak points."
Joanna Vodopivec Prabhas PokharelOne Research Team for All - Influence Without Authority
March 9, 2022
"Some participants prefer tasks to be direct and focused, without lengthy scenario explanations."
Rebecca ToppsPlanning and conducting remote usability studies for accessibility
September 10, 2020
"About 20% of users shown our Net Promoter Score survey provide written feedback that can be analyzed for improvements."
Alan Williams Rose DeebDesigning essential financial services for those in need
February 10, 2022
"If Charlie has been tampered with, Alice needs a clear way to be alerted that she shouldn't trust it."
Heidi TrostWhen AI Becomes the User’s Point Person—and Point of Failure
August 7, 2025
