I've spent 8 years designing for humans navigating complex systems. The most interesting ones now act on their behalf.

Agentic AI is the design problem I care most about right now. When a system can take real actions, budget shifts, campaign launches, data writes, the interface has to do more than present information. It has to make the agent's intent visible, its confidence honest, and its failures recoverable. I've built that from scratch.

My process starts with understanding the system before touching the interface. I care about the logic underneath: the flows, the states, the edge cases, because that's where most design decisions actually get made. For agentic systems that means mapping decision points, intervention surfaces, and failure modes before a single screen exists. The visual layer follows. It doesn't lead.

Before AI, I spent years in B2B SaaS, fintech, and merchant tools. Products for users under real constraints: low bandwidth, high cognitive load, no time for onboarding. That background makes me useful in AI product work. The discipline is the same. The stakes are higher.

I shoot film. Good espresso matters. Mountains reset the pace. All of it shapes how I think about constraint, precision, and what gets left out. 📷 ☕ ⛰️