The things I make don’t look much alike.
A brand, a way to learn how AI works, a paper. On the surface they have little in common. Underneath, each one started as a question I couldn’t put down.
I started LuMiNx around a belief I keep coming back to: Reach the Unreachable.
To me it means choosing the questions that seem too far away, and staying with them long enough to find out how close they really are.
I also shape how LuMiNx looks and moves, from its identity to the website itself. I’ve never thought of that as a separate job: the way something looks is part of what it believes.
Vector.D is a project I’m building to explain how large language models and AI agents work.
Most explanations of AI hand you terms and formulas. Vector.D tries to show the mechanism underneath, so you can see why a model behaves the way it does.
Once the pieces connect, the terms stop being something to remember. They simply make sense.
Learning should feel less like memorizing a manual and more like discovering how something works.
Most questions end up as thoughts. A few keep pulling at me until they turn into research.