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
"Systems thinking is not just about complexity but grounded in ethics to help deliver a better future for humans and the planet."
Sheryl CababaExpanding your Design Lens with Systems Thinking
March 28, 2023
"We talked all about resilience, curated by myself yesterday."
Bria AlexanderDay 3 Welcome
September 25, 2024
"There is no such thing as one-size-fits-all methods; what works depends on the component’s evolutionary stage."
Simon WardleyMaps and Topographical Intelligence
January 31, 2019
"Some of the older content has discoverability problems, which is typical with blogs."
Jorge Arango[Demo] How to re-categorize content at scale using LLMs
June 5, 2024
"Innovation is not a spice that you sprinkle on and suddenly magical new things come out."
Jeff GothelfInnovation Studios: the Engines of Enterprise Experimentation
May 14, 2015
"Creating an environment where people feel safe to try design helps build enthusiasm and psychological safety."
Lona MooreScaling Design Beyond Designers
June 11, 2021
"Complexity is not a barrier to accessibility; even complicated games like The Last of Us are accessible."
Samuel ProulxInvisible barriers: Why accessible service design can’t be an afterthought
December 3, 2024
"You have to keep the injection sites moving because of scar tissue, adding to the body hacking complexity."
Daniel J. RosenbergDigital Medicine Design
September 26, 2019
"Design systems allow us to wrap up a prototype quickly, test, fail fast, and iterate."
Scher Foord Corey Greenltch Sarah RoweTurn the Ship Around: How to Apply Design Thinking Across Your Organization
June 10, 2021
