Peter Van Dijck
Founding Partner and CEO, The Helpful Intelligence Company
Early in his career, Peter founded three startups, and was the author of a book on Information Architecture. More recently, he built a 500+ remote engineering, design and product team in Latin America, for clients like Google, Apple and Goldman Sachs. Peter is passionate about making the world better with digital products, AI and working on the climate emergency. Peter runs modelcontextexperience.com to help product people, UX designers and others learn how the best AI products are built.
Rosenverse talks by Peter:
" You can’t just be a designer adapting UX workflows assuming everybody else’s workflow stays the same; everything is changing with AI. "
Hands on AI #3: Claude Code for UX people
October 22, 2025
" Model companies are students in a classroom wanting good points—they’re happy to run external expert evals to improve. "
Hands-on AI #2: Understanding evals: LLM as a Judge
October 15, 2025
" Confidence scores from the model are not a good idea because the model has no memory. "
Hands-on AI #1: Let’s write your first AI eval
October 8, 2025
" Books are semantically very rich and thus pull content much more effectively in our vector database than conference talks. "
GenAI for UXers: A Rosenbot Demo and Discussion
September 11, 2025
" Tool use lets the model decide itself when to call external APIs or services to get needed information. "
Building impactful AI products for design and product leaders, Part 3: Understand AI architectures: RAG, Agents, Oh My!
July 30, 2025
" It’s much easier to classify an answer than to generate an answer, and that’s why LLM as a judge works. "
Building impactful AI products for design and product leaders, Part 2: Evals are your moat
July 23, 2025
" Synthetic data is exceedingly useful because you can generate it at scale for training and evals. "
Building impactful AI products for design and product leaders, Part 1: The new product journey
July 16, 2025
" Building evals is like building tests to know if your AI system is good enough, and you can rerun them as models improve. "
Designing AI-first products on top of a rapidly evolving technology
June 10, 2025
" The product is imperfect, probably about 60 percent there, but if we get to 80 percent it will be a compelling tool for UX content. "
Coffee with Lou #4: Taking a Peek Under the Rosenbot's Hood
June 14, 2024
" We invented the ‘prompt deep dive’ technique to spend lots of time on one prompt to deeply understand it. "
Building the Rosenbot
June 4, 2024