barely discernible

I work at the seam between mathematics, computation, and design. I trained as a computer science engineer, took a long detour through design school, and spent most of the last decade teaching: mathematics, algorithms, and human-computer interaction. The detour is the point. It taught me to distrust the gap between a clean result and a thing someone actually has to use.

Three commitments organise the work. Mathematics I treat as a language for reading the world and for designing inside it, less a store of answers than a way of seeing structure. Computation is how that seeing becomes leverage, the interface that turns an idea into something you can run, test, and hand to someone else. Ethics is the part that refuses to be optional: a habit carried over from human-centered design, and a long argument with technology, of asking who a system is really for and what it quietly assumes.

Lately that has meant machine learning and image processing aimed at engineering problems, and a stubborn interest in the choices made upstream of any model: how a physical thing becomes data, and what that representation lets a model see or keeps it from seeing. I care that the tools stay open and legible rather than locked away. I am not the quickest to an answer, but I am good at finding the question worth asking, and patient enough to stay with it.