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
"Culture is the lens through which we experience our lives and products today."
Neil BarrieWidening the Aperture: The Case for Taking a Broader Lens to the Dialogue between Products and Culture
March 25, 2024
"If your users are happy, that anecdotal feedback often carries more weight early than quantitative metrics."
Lada Gorlenko Sharbani Dhar Sébastien Malo Rob Mitzel Ivana Ng Michal Anne RogondinoTheme 1: Discussion
January 8, 2024
"If you say great user experience, no, it’s not specific, it’s not measurable, it’s not actionable."
Fredrik MathesonFirst-time users, longtime strategies: Why Parkinson’s Law is making you less effective at work – and how to design a fix.
June 8, 2016
"The AI planning monkey and banana problem made me question whether getting the bananas is really the end goal."
Liwei DaiThe Heart and Brain of the AI Research
March 31, 2020
"Each team should talk about their core values like integrity, accountability, and respect to work together with the same mindset."
Jon Fukuda Amy Evans Ignacio Martinez Joe MeersmanThe Big Question about Innovation: A Panel Discussion
September 25, 2024
"Automation reminders to authors help close the loop on what actions come from their published insights."
Michelle Bejian Lotia Anne-Marie MorellRolling Out a Repository: How Zapier Centralizes Insights from Across their Organization
March 28, 2023
"Innovators need a tribe to recharge their batteries because you get no from every direction."
Kristen Honey"Let’s Talk About Data and Crisis”: Public Digital Service Delivery = Open Data + Human Centered Design
November 18, 2021
"Speculative use cases connect technical components to human impact and help prioritize work into slices of value."
Lais de Almeida Maria IzquierdoDesigning Data Services
December 4, 2024
"The highest leverage thing you can do is design statements that capture the essence in ways that set sail a thousand ships."
John CutlerThe Alignment Trap
November 29, 2023
Latest Books All books
Dig deeper with the Rosenbot
What are effective workflows for integrating 3D modeling and AI image generation in design projects?
How does AI help maintain UI consistency across multiple development teams in large enterprises?
How can public sector design teams implement knowledge management to mitigate knowledge loss during staff turnover?