Portfolio

Selected work · 2003 — 2025

Four projects across two decades and a few unforgiving industries — design systems for satellite ground control, kiosk experiences, fifty-plus congressional websites, and an e-commerce overhaul on top of four point-of-sale systems. The pattern, usually, was the same: untangle the constraints first, then design what you can actually build.

Astro UXDS

The bespoke design system commissioned by the U.S. Air Force and Space Force to support ground control for satellite operations — and the only one of its kind to outlive its program of record.

I joined Astro at the start as Lead Design Engineer. By version 2.0 I owned the program. The brief was narrow — components for satellite ground stations — but the work that mattered was wider: convincing a defense organization that an open, public, framework-agnostic system would outlast the access-controlled alternatives that had failed before it.

Three early decisions made Astro durable. I open-sourced the documentation and components, which sounds obvious until you’ve watched a year disappear into clearance reviews for a private wiki. I migrated the design assets from Sketch to Figma — Macs are rare in the Space Force, and asking operators to install software they couldn’t get approved was a non-starter. And I bet on W3C Web Components, knowing our consumers used a half-dozen frameworks and that some of these systems would still be running in 2040.

Over the program’s life I directed three migrations — Sketch to Figma, Lit to Stencil, and a heavyweight CMS to docs-as-code in markdown — without breaking a release. The work that surprised me, and probably mattered most, was the pattern library: a Timeline component used across satellite scheduling, classification markings to MIL-STD-1472, status systems, alerts, and full sample applications grounded in user research with operators.

Client: U.S. Air Force / Space Force
Role: Lead Design Engineer → Program Owner
Years: 2015 — 2025
Adopting organizations across government, F500, startups: 50+
Component downloads at peak: 30K/mo
Documented dev-time savings on adopting projects: 30%

Astro UXDS marketing site home page — 01 Astro UXDS Docs-as-Code site — the public front door that opened access for teams building applications for the space domain. astrouxds.com

Astro Storybook component playground — 02 Early adopted of Storybook targeting developers and supplementing behavioral and design guidance. astro-components.netlify.app

Astro Figma library — 03 Figma library — adoption skyrockted after migrating to Figma. Transition required detailed planning and coordination with project teams. figma.com/@astrouxds

Astro Timeline pattern in a scheduling app — 04 Timeline pattern in a working scheduling app — a unique component for satellite ops with full guidance, web components and Figma parity.

Astro sample applications — 05 Sample applications and case studies — built from operator engagement sessions to show how components compose.

drag, scroll, or swipe

Highway to the Future

A traveling exhibit of 18 interactive touchscreen experiences, designed to teach hybrid drivetrain concepts to consumers who’d never thought about a car powertrain in their lives.

Toyota was launching the second-generation Prius and needed a way to explain Hybrid Synergy Drive to people standing in a parking lot. The answer was two semi-trucks of touchscreens — 9-inch up to 42-inch — and a barcode scanner at every station so a single user’s path through the experience earned them survey-eligible reward points.

The brief sat in an awkward middle: more substantial than a trade-show demo, less interactive than a game. Wireframes were drafted in OmniGraffle and the production work shipped in Adobe Flash, which was the right tool for full-bleed touchscreen in 2003 even if it ages poorly in screenshot form.

What I’d flag now: this work predates the discipline of measuring exhibit dwell time properly. We had scan data and survey responses but no good way to know which of the 18 screens actually moved the needle on consumer understanding.

Client: Toyota
Role: Interaction & Visual Design
Years: 2003 — 2004

Toyota Mobile Hybrid Experience hero — 01 Hero screen for the Mobile Hybrid Experience — and the small-screen variants that toured with the trailers.

02 Wireframes-to-finished comparison across three of the 18 experiences.

drag, scroll, or swipe

United States Senate

Between 2004 and 2015 I designed more than fifty front-facing congressional websites and a handful of internal applications — for clients with no budget for user interviews and a deep institutional preference for the way things were done last time.

Government projects of this scale rarely fund market research, let alone qualitative interviews. So I worked with what was free and abundant: six years of Google Analytics across dozens of senator and committee sites, plus the on-site search logs.

I overlaid click-rates on the home page, surfaced top-time-on-page content, and pulled the searches that returned no results — which is usually the most honest signal a website has about what visitors actually want. With that data I made a case the client could argue with: a radically restructured layout that gave 70% of the home-page real estate to the content 70% of visitors actually came for, against a more traditional layout that respected the client’s communications priorities.

The traditional layout won. Government is conservative; that was the right outcome for the institution if not for the optimization curve. The exercise still mattered — every project after it started from a research brief, not an aesthetic preference.

Client: United States Senate
Role: Designer & Researcher
Years: 2004 — 2015

Senate analytics research deck — 01 Analytics research deck — top topics, hidden content, time-on-page, and entrance pages distilled into a one-page argument.

Two competing Senate site wireframes — 02 Two competing wireframes: the data-led layout (left) and the more traditional layout that ultimately shipped (right).

Sketch to wireframe to mockup to HTML progression — 03 Sketch → wireframe → mockup → HTML. The full progression for one of the 50+ sites.

drag, scroll, or swipe

zPizza

A complete redesign of the zPizza web presence and a bespoke online ordering system that had to sit gracefully on top of four different point-of-sale systems and three e-commerce vendors.

zPizza came to us with the kind of integration problem that gets quietly worse every quarter — four POS systems, three e-commerce vendors, frustrated franchisees, and customers who’d taught themselves to call the store instead of using the website. The brief was “a redesign.” The actual job was “untangle this.”

I led with research. A competitive teardown traced the full transaction across three direct competitors and two best-in-class e-commerce comps, on desktop and mobile, with annotated pros, cons, and lessons. A 5,000-respondent customer survey, paired with a separate franchisee survey segmented by POS and vendor, told me where customer pain and operator pain overlapped — and that overlap was the prioritization map.

The most useful single finding: 87% of users were on a browser less than a year old, and nearly half were already on mobile. That gave me license to lead with mobile in wireframes — unusual for a pizza chain in 2013 — and grow tablet and desktop out of the strongest mobile direction. The full ordering flow was wireframed end-to-end and built as a working prototype in Node, Backbone, Express, HTML5, and jQuery.

Client: zPizza
Role: Lead UX & Research
Years: 2013
Customers surveyed: 5,000+
POS systems unified behind one ordering UI: 4 → 1
Users on browsers < 12 months old: 87%

zPizza competitive analysis and customer survey — 01 Competitive analysis and customer survey — the research foundation that shaped every later decision.

zPizza customer and franchisee issue synthesis — 02 Synthesis: customer issues, franchisee issues, and the overlap that became the prioritization map.

zPizza end-to-end user flow — 03 End-to-end ordering user flow — geolocation, store selection, delivery boundaries, cart persistence. Pioneering features at the time later adopted by larger players in the space.

zPizza mobile ordering wireframes — 04 zPizza’s target market were early adopters and had a larger mobile share of visitors than competitors at the time, the site and platform was designed as mobile-first before mobile-first was widely adopted.

zPizza mobile mockups — 05 Mobile mockups in the chosen visual direction.

zPizza tablet mockup — 06 Tablet — grown out of the selected mobile direction.

zPizza desktop mockup — 07 Desktop — the same system at the largest viewport.

drag, scroll, or swipe