I design and ship AI-powered products for complex business environments.
Most job searches only start once a role is already posted — and already competitive. I designed and built a live n8n system that ranks companies by hiring-momentum signal across news, Reddit, and LinkedIn, surfacing real opportunities before they go public.
Creators have goals but no clear path from idea to action. I researched, prototyped in Figma, and built (React, with Claude Code) an AI co-pilot that turns a stated goal into a concrete plan — a self-directed capstone proving I can own a product end-to-end, AI included.
Opportunity Radar found the right companies but left the harder calls unsolved — is this worth pursuing, how do I walk in prepared, what do I ask for. Two agents and one honest pipeline close that gap: fit verdicts, interview prep, and offer guidance in one flow.
Underwriters were working across a mash-up of legacy tools with no shared system. I led design for a single modern underwriting platform — partnering directly with engineering to close design-system gaps and cut rework — built for a regulated, high-stakes B2B environment.
Program teams tracking poverty-reduction work lacked shared visibility across field visits, participant selection, and multi-level reporting. I scoped and led UX for dashboards and decision-support workflows that helped distribute $10M+ in program funds across program organizers, branch managers, and regional teams.
A portfolio claiming an AI-native workflow needs to prove it, not just say it. This site is that proof — designed and built using AI-assisted workflows across Figma, Claude, and Cursor, documented as it happens.