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
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
"Computers got ten years better overnight with the arrival of GPT-3 in November 2022."
Matt WebbContext Window: Five Futures for AI
June 11, 2025
"It’s about helping your colleagues be more productive and make better decisions to help customers."
Sofia QuinteroBeyond Tools: The Messy Business of Implementing Research Repositories
March 10, 2022
"Separating genders in research can perpetuate stereotypes rather than challenge them."
Mansi GuptaAdvancing the Inclusion of Womxn in Research Practices
September 15, 2022
"Data per se are not the problem; it is the way we are using the data that can be problematic."
Matteo GrattonCan Data and Ethics Live Together?
October 1, 2021
"People were more likely to explain their thought process and just more context around the particular situation of their org."
Tara TresselInvestigating qualitative depth of AI-moderated interviews
March 10, 2026
"Climate UX feels like 1995 again, an exciting time of discovery and cross-domain jumps."
Louis Rosenfeld Matt Jones Olga Khroustaleva Michael Leggett Karol MunozDo you want to work on climate? (Climate UX Discussion Series)
November 15, 2023
"A lot of people were just going and downloading the tear sheet, which is to be expected."
David CroninThe GE Design System and Thoughts about Craft at Scale
May 13, 2015
"We decided not to build a new app but instead integrate within the legacy electronic health record system."
Laureen Kattan Julie KimCentering Patients and Clinicians in a Complex Government Ecosystem
November 29, 2023
"If a website is hard to use, motivation and approach drops, and users are less likely to engage or convert."
Michael WeirMixed Methods and Behavioural Science
May 26, 2023