Design to ship.
Build for need.
I coach and lead teams to set up the workflows that ship what people, and their AI agents, need. Every build starts from UX research and analytics, so nothing ships on a hunch.
Book a 25-minute introWorked with
- McGraw-Hill
- The Economist
- CBS · EcoMedia
- EMC
- IEEE
- World Vision USA
- Junior Achievement
- Bell Works
- AeroTorch
- Advanced Companions
Scroll · three recent projects, then back to 1997
A torch maker, selling online for the first time.
- Problem
- Product knowledge lived in a paper catalog and the owner’s head. No way to sell direct.
- Action
- One product database first, then a store, search and feeds on top, built with AI agents inside checks I set.
- Solution
- A site and database as the cornerstone of its digital transformation. First sale the day the store opened.
- 4 channelsOne catalog feeds the site, the store, a distributor’s import, and Google and AI shopping feeds
- 423 specsAnd 356 links across 70 products, modeled once
- 7 weeksFiling cabinet to working catalog
For your team
The groundwork-first pattern a platform team uses, at main-street scale.
Built in the field, in between visits.
- Problem
- Independent pet care pros run on texts, notebooks and Venmo. When a pet passes from one pro to the next, walker to trainer, its care history stays behind.
- Action
- Product lead, designer and, with AI agents in a harness I set up, the engineering team.
- Solution
- Families paying real invoices since March 2026, with Lulu keeping the care record.
- 32 daysFirst commit to first paid invoice
- 15Production releases in eight weeks
- 22 rulesWritten from 22 logged incidents
For your team
An AI-first product from zero to paying customers in 32 days, with guardrails that let it keep shipping safely as it grows.
Harnessing off-the-shelf GenAI
- Problem
- Every good AI result was a one-off nobody could repeat.
- Action
- Tested what off-the-shelf tools could really do, then built the context they lacked: specs, examples, critique.
- Solution
- A 17-person team using GenAI that holds up in anyone’s hands.
- 4–5More runs of the spec, then a write-up for other teams
- On stageThe artist-appeal backdrop
- Snap ShotsThe war room, still running
For your team
You don’t need a custom model to get dependable AI work. You need the right context around the tools you already pay for.
Earlier work · back in time
Before these three, six eras.
- 01Manual design systems2018 to about 2022Goldman Sachs with TandemSeven, ZS, MRM for Sunovion
- 02Early agentic AI2015 to 2020IEEE, Junior Achievement (JA MyWay), metaMeJA MyWay · metaMe
- 03Expanding outside of digital2014 to 2019Bell Works and coworking, Meetup, teaching at Parsons and Rutgers
- 04Mobile first2011 to 2016Wisdom & Craft, PEOSpectrum
- 05Web design · IxD · IA · usability1997 to 2011Independent UX consulting; the essays from 2007 on
- 06Early daysBefore 1997Keyboard, joystick, music, art and writing
How I work
Six Ds in three loops, with Anthropic’s 4Ds for working alongside AI.
About
UX and product leader since 1997. I design AI products and ship them to production. I’ve taught at Parsons, Rutgers and NYU.
More about meContact
Working on something like this?