Writing

Notes on how people decide things.

Most of what I write about sits underneath the interface. It is unorthodox for someone in design to take a less visual approach, but I find that writing freely helps me “sort things out”. I suppose you could say this is the Feynman Technique in practice.

September 19, 2026 Agentic design Reaching the Summit My side project needed a parts catalog nobody sells, so a crawler builds it in runs that last for days. Summit is the instrument I built to steer those runs: every instruction a memory symbol, every symbol at an altitude, and a marker you drag to change the run's mind without stopping it. September 6, 2026 AI design principles Altitude, Confidence, and Consequence I published eight posts thinking they were eight different arguments. They weren't. This is the shape they were circling, assembled backwards, with the weak parts flagged. August 24, 2026 AI design principles The LLM Black Box and How It Affects Design Most of us now work with a model every day and have stopped thinking about what we put into it. That's a design problem as much as an engineering one. August 20, 2026 Part 2 of 2 Decision-Making in UX: The Two-System Concept, Part 2 Part one made the case for designing toward recognition. This one is about how you find the places you aren't, by putting a clock on the goal and reading where the time goes. July 16, 2026 Part 1 of 2 Decision-Making in UX: The Two-System Concept, Part 1 Kahneman calls intuition nothing more and nothing less than recognition. Take that seriously and most of what we call good design turns into something you can design on purpose. June 11, 2026 Product design Why Memory Sequences Matter for Product Design Knowledge is memories wired together. They store in order, they replay in order, and when you change one step in the middle of a sequence somebody has learned, you break more than a screen. May 14, 2026 Cognition Hierarchical Thinking and Its Impacts on AI Design Kurzweil argues the neocortex is a stack of pattern recognizers talking only to their neighbors. Sit with that for a while and a lot of what feels strange about working with a language model stops feeling strange. April 16, 2026 AI design principles Trust Calibration, Not Trust Maximization We keep treating trust as a number to push upward. Somebody who trusts a system more than it deserves is a failure, and so is somebody who has stopped using it. March 19, 2026 Craft Why AI-Generated Design Looks Generic, and It Isn't the Model’s Fault Ask for a clean modern landing page and you get the mean of the internet. The model answered the request you gave it, and the request had almost nothing in it. February 12, 2026 Marketing × design What Happens to the Funnel When the Agent Does the Browsing We have spent twenty years designing pages for a person who scans, hesitates, and can be persuaded. A growing share of the traffic now does none of those three things.