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 #2: Understanding evals: LLM as a Judge
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 second talk in the series, Peter Van Dijck of the helpful intelligence company will show you how to create an eval for an AI product using an LLM as a judge (when we use a Large Language Model to evaluate the output of another Large Language Model). We’ll have a look at how that works, but also dig into why this even works. Are we creating problems for ourselves when we let an LLM judge itself? This talk is hands on; and there will be plenty of time for questions. You will go away understanding when and how to use LLM as a judge, and build 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 are a foundational feedback loop defining what 'good' means for AI products, helping to measure and improve systems continuously.
-
•
Evaluating fuzzy, subjective AI outputs requires innovative approaches such as using LLMs as judges to score results.
-
•
Binary (yes/no) scoring is more reliable than rating scales with ranges because LLMs lack internal memory and consistency.
-
•
Starting evals early (week one of a project) drastically improves AI product outcomes, but many teams delay due to perceived complexity.
-
•
High-risk or important tasks should be prioritized for evals instead of attempting broad coverage.
-
•
Assigning a dedicated owner or 'benevolent dictator' for evals who works closely with domain experts accelerates feedback and quality.
-
•
Creating a written constitution of principles helps concretize AI behavior goals and guides prompt and model training.
-
•
Most current eval tooling is too technical, slowing iteration cycles and making expert involvement inefficient.
-
•
Custom feedback interfaces tailored to expert users significantly speed up evaluating AI outputs in domains like healthcare and law.
-
•
Diverse perspectives from UX, product, strategy, and domain experts are critical in defining and refining what 'good' means in AI systems.
Notable Quotes
"Evals are everywhere, right? Everybody's talking about evals. It is like one of the key things in developing useful AI products."
"You want to ask an LLM to evaluate the fuzzy stuff because there’s no black and white output."
"LLMs don’t have memory, so rating on a scale from one to five is pretty random. Better to have yes or no answers."
"One of the biggest problems in AI building is evolving your prompts and having a fast feedback loop."
"By starting to categorize risk in detail, you naturally lead to better prompts and better evals."
"A constitution is a very good exercise: write down your system’s principles and values to help guide its behavior."
"Use custom systems for experts to quickly review and rate outputs, making feedback cycles much faster."
"Evals define a shared definition of good with tests to measure it, and that is the secret sauce for building great AI products."
"Model companies are students in a classroom wanting good points—they’re happy to run external expert evals to improve."
"The more I work with evals, the more I think UX and product people need to be involved because of the need for diverse perspectives."
Or choose a question:
More Videos
"Trauma is a response to anything that's overwhelming—too much, too fast, too soon, or too long coupled with a lack of protection or support."
Matt Bernius Sarah Fathallah Hera Hussain Jessica Zéroual-KaraTrauma-informed Research: A Panel Discussion
October 7, 2021
"Storytelling and facilitation are becoming more important than just hard research skills for driving strategic product decisions."
Jen Cardello Dr. Shadi Janansefat Alex WrightCurating insight: Strategies for integrating knowledge across research functions
March 11, 2025
"We come to these spaces not to escape but to partake of them."
Sha HwangThe First Fifty Years of Civic Design
November 16, 2022
"Navigating complexity is like sailing in open ocean. A GPS won’t help if you don’t know how to read wind and water."
Kyle GodbeyNon-linear service design for complex adaptive systems
December 10, 2025
"Measuring ops impact by how much people are using the research and embedding feedback loops is much more valuable than just looking at money saved."
Brigette MetzlerScaling ResearchOps: Helping Researchers do Their Best Work
March 30, 2020
"Healthcare is not a silver bullet for AI; a lot of context comes from non-verbal cues where generative AI doesn’t apply."
Jennifer KongJourneying toward AI-assisted documentation in healthcare
June 5, 2024
"It was really hard to get senior stakeholder attention when starting late, especially when the topic is not fashionable anymore."
Magdalena ZadaraZero Hour: How to Get Far Quickly When Starting Your Digital Service Unit Late
November 16, 2022
"AI has eaten the world and will continue to eat the world forever."
Husani OakleyTheme Three Intro
June 6, 2023
"People stay in roles mostly because of the people and the relationships they have on their teams."
DesignOps and The Great Talent War of 2021
August 19, 2021
