I design AI products.
Then I ship them.
13+ years at Google, BMW, and enterprise fintech — then I designed and engineered aiOS end to end, solo: the design system, the codebase, the agent architecture. One hand, zero handoffs.
Watch what shipped
The full walkthrough in four chapters — skip around like stories. Everything shown is live in production, narrated by me.
My design philosophy
I believe AI should amplify human capability, not replace it. My career has been one continuous thesis: software should feel like a conversation, not a form. I prototyped conversational banking at Lumin, built design foundations for AI products at Google — and then went and built the whole operating system myself.
“It's less important what AI can do. It's more important how it makes people feel.”
What good UX feels like
A real conversation, rendered with the production iMessage engine I built.
Two texts in. A morning of work out.
This is the whole product thesis in one thread: the owner sends two short texts, and a morning of work happens — leads qualified, a follow-up drafted for approval, content queued and posted. The human approves; the machine executes.
No dashboard. No prompt engineering. The most-used surface of aiOS is the one every business owner already checks eighty times a day.
Human judgment, AI Execution
Six digital employees, one brain — the live component, straight from the aiOS production codebase.
Case Studies
Four products, one thesis: complex AI that feels simple.
aiOS — an AI employee that runs your business
Founded, designed, and coded an agentic AI platform spanning iMessage, web, and voice — live in production with enterprise deployments. The RTR Vehicles deployment resolves 92% of support conversations automatically.
AI-powered banking for 127 credit unions
Conversational onboarding, AI design companions, and an enterprise platform serving 8M+ members — replacing form-wizard banking UX with natural conversation.
Design foundations for AI at scale
Unified five ML/AI product teams onto one design system — 67 foundation components with 85% adoption, built for AI capabilities that didn't exist yet.
Smart EV charging for a smarter grid
BMW's first dynamic charging experience — ML-scheduled charging that cut peak-hour load 87% while guaranteeing a full battery by morning.