12+ years building things people use — design systems, creative tools, AI agent systems, and production UI. Currently at Epsilon leading AI-powered creative tooling. Independently shipping products with AI and building autonomous agent systems that think, execute, and ship.
Leading a team building no-code creative ad production tools. Architecting AI features: headless automated pipelines and user-facing AI that transforms images into layers and converts natural language into layout edits. Thousands of ad variations from flexible templates.
Grew Epsilon's component library from a side project into the core design system adopted across products. Founded the UX engineering discipline, built the team, wrote the documentation, and drove org-wide adoption. Championed WCAG accessibility as a design constraint embedded into every component.
AI photo style transfer platform. 40+ comedy styles, 230+ tests, config-driven architecture, Lighthouse 95+. Built with Next.js, TypeScript, and OpenAI APIs. Live at makemea.ai
makemea.ai →An agent skill that discovers and resolves open source GitHub issues autonomously. Finds issues, forks repos, implements fixes, and opens PRs with full AI disclosure. Built for the OpenClaw ecosystem — published on ClaWHub.
clawhub.com →A shared living canvas where AI agents create art through marks. Dual-layer rendering: live dot primitives plus AI-painted compositions that evolve continuously. Stripe payments, BYOC agent API, epoch timeline with replay.
sprawl.place →Visual regression testing pipeline for AI agents. Captures baseline screenshots with Playwright, pixel-diffs against new builds, and gates deployments on similarity thresholds. Published on ClawHub.
clawhub.com →Large-format digital experiences for major brands. Personalized animated photo experience for Coca-Cola (malls across US). Award-winning motion-activated Photo Booth for PepsiCo (Kinect gesture recognition). 3D interactive experience for Accenture at JFK Airport.