What I do, where I've done it, and the work that came out of it.
I write the documents work actually gets built from. Requirements, PRDs, roadmaps, sprint plans, and the architecture calls sitting underneath them. Most of it starts with a founder or partner session that I run myself, and ends as tickets a team can pick up on Monday morning.
I came into product through design, so I still do the work. Interfaces for AI products on web and mobile, design systems, brand identity, and workflow diagrams. When something needs to exist quickly rather than perfectly, I build it in Framer or Webflow instead of writing a brief for it.
I write my own tooling. Custom Claude skills, chained agent pipelines, prompt and evaluation design, and MCP and API integrations. Most of it exists because the workflow I needed did not, and the fastest way to get it was to build it myself.
A document leaves too much open. So I stand up a working version early with Claude Code and Codex: the real user flows, the real states, and the whole platform structure, running on mock data. Clients and their users can move through it and see how the product will actually operate before anyone commits a sprint to building it.
Before anything gets built I do the unglamorous part. Market and competitor research, customer segmentation, prospect evaluation, and technical deep dives to work out whether a partner and an idea are genuinely a fit. The positioning and the copy usually end up being mine too.
Most of my work is agentic AI applied to regulated or operationally messy industries. Healthcare and medical devices, fintech and capital markets, real estate, freight, construction, and edtech. The pattern repeats: high stakes, scattered data, and people who do not want another dashboard.
Senior Product Manager & Designer
Building infrastructure for AI agents.
A conversational AI layer for medical practices covering both front office work, scheduling and intake, and back office work, claims, prior authorization, and denials. I turned a founder deep-dive into the product brief, mapped every use case against our build library, and named the two-product wedge the team qualified.
Patients forget most of what they are told after surgery. A clinician uploads a recovery protocol and the system extracts the timeline, writes a script, clones the clinician's voice, and delivers it as a narrated avatar video with check-in questions. I defined the generation flow and wrote the spec.
An equities research platform run by autonomous analyst agents. They pull earnings filings and PDFs, monitor news, and produce analyst-grade coverage, alongside a matching layer for private share demand. I ran the founder sessions, extracted the use cases, and shaped them into a scoped requirements set for the PRD.
Buyers make the largest purchase of their life on a listing page and a gut feel. Paste a property link and this returns true monthly cost, a fair price range, and seventeen risk factors, each with its source and confidence. I designed the scoring architecture and tiered report structure.
Every company has a process on paper and a different one in practice, bent around emails, spreadsheets, and workarounds. This engine reconstructs how work actually flows from system logs and unstructured artifacts, then shows the bottlenecks and rework. I built the framework spec and its first vertical module for freight.
Public funding goes unclaimed because chasing it is a full-time job nobody staffs. The engine searches a large grant corpus, filters hard eligibility rules, then ranks and explains the remaining matches with draft proposals attached. I owned the roadmap, phased the matching pipeline, and set up the delivery tracker.
Client documents used to become requirements, a PRD, a roadmap, and tickets through slow manual work. I wrote four connected agent skills, a requirements analyzer, PRD generator, roadmap generator, and tracker sync, then chained them into one pipeline with human checks between stages. Cycle time dropped over seventy percent.
An open-source multi-agent system that takes an open-ended task, plans it, then browses and executes across tools until it returns a finished result. It has 450+ stars and 75+ forks. I led product and design, defining agent handoffs, what users see during long runs, and the interface.
Associate Product Designer
Turbocharged GenAI fine-tuning & inference platform.
