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
"Most science starts with somebody pulling a number out of their ass — it’s okay to start with a gut instinct or guess."
Caroline Jarrett Erin WeigelHave fun with statistics?
December 12, 2024
"We have to step outside research and understand AI, technology, and product processes to put ourselves in the right space."
Megan Blocker Lada Gorlenko Fatimah Richmond Molly StevensWhat UX research maturity looks like and how we get there [Advancing Research Community Workshop Series]
November 9, 2023
"Treat unemployment as a full-time job; be relentless and turn over every stone to improve your chances."
Corey LongHiring in DesignOps: A Critical Study on How to Hire and Get Hired
September 23, 2024
"You have to find your champion as high up in the company as possible to push for research and equitable pay."
Mike DavidsonFireside Chat
March 11, 2022
"Being a chief of staff means remembering and helping other people get their shit done."
Lisa GirondaOpener: Chief of Staff–An unexpected journey
January 8, 2024
"Context retention is tough; Siri and Google Assistant completely lose prior contexts after one action."
Greg NudelmanDesigning Conversational Interfaces
November 14, 2019
"Keeping communication fluid with participants until the moderated session keeps them engaged and less likely to forget."
Roberta Dombrowski Lianna Aduana5 Reasons to Bring your Recruiting in House
September 30, 2021
"The AI assistant enables ongoing conversation and follow-up questions to refine persona insights dynamically."
Gillian Salerno-Rebic Mark MicheliFrom Insight to Impact: How JourneySpark Used WEVO Pulse + Pro to Drive a 50% Lift in Ad Engagement
June 11, 2025
"Every hiccup in a process and every outdated protocol accumulates a toll on the team as well as financial costs."
John Calhoun Rachel PosmanMeters, Miles, and Madness: New Frameworks to Measure the (Elusive) Value of DesignOps
September 24, 2024