Enrico Piovesan

Software should work for users, not for the runtime it happens to run on.

I spent years building products across startups in travel, education, and payments. Good teams, real users, hard deadlines. And the same problems kept showing up regardless of the company or the stack. Business logic rewritten for every new runtime. Systems that worked in isolation and fell apart at the boundaries. Codebases that only made sense to the people who built them, and only for as long as those people stayed.

At some point I stopped treating those as facts of life and started asking why they kept happening. The answer, most of the time, was that the architecture had been designed for the environment instead of the problem. The runtime got more respect than the user.

That question is where everything came from. UMA started as an answer to portability. Why should the same business logic exist in five different forms across browser, backend, edge, and AI pipeline? C-DAD started as an answer to navigability. Why can AI agents describe what software does but not why it does it? Traverse is the runtime that takes those ideas further. Each one came from a real problem that kept showing up, not from a whiteboard.

Two books, five research papers, and four open source projects later, the conviction has not changed. It has just gotten more specific.

I am a Platform Software Architect at Autodesk. Before that I spent four years as a Domain Software Architect at Ford, and before that I co-founded companies in travel, education, and payments. I moved from Italy to Canada about ten years ago and ended up in Golden, BC, which is a small town in the Purcell Mountains with exceptional snow. I am a certified ski instructor and a ski patroller. I have two daughters who think the word monolith is scary, and I have not corrected them. I still make real carbonara. No cream, no shortcuts.