Clayton Young
01 · Work

Shipped, not mocked.

Everything below is live — you can use each one right now. Three are shipped products with real users; one is a working study, and it's labeled as one. For each: the problem, the design decisions, and the outcome. I'm the designer and the person who built it — so nothing here died in handoff.

01

Market Pulse

Live demo ↗
A prediction-market interface study — live build, simulated data · Next.js · Framer Motion

Prediction-market UIs drown people in numbers that keep changing. I wanted the price to feel alive without ever feeling anxious — the difference between a market you watch and one you close.

Design decisions
  • Prices glide on springs instead of snapping — movement reads as information, not alarm.
  • Tabular numerals everywhere a digit changes, so nothing shifts or jitters.
  • One market gets the hero treatment; the rest recede — hierarchy over density.
  • Green/red carry direction only — never decoration — so color stays trustworthy.
Honest label: a working study, not a shipped product — real interface, simulated prices. Built on my own nights before I ever saw this posting.
02

Sip: Drinks Abroad

App Store ↗
Consumer iOS app, shipped to the U.S. App Store · React + Capacitor

Standing at a bar in a country where you can't read the menu is a small, real consumer problem. The product had to work instantly, offline, in loud rooms, one-handed.

Design decisions
  • Fully offline-first — the moment of need has no Wi-Fi, so the design assumes none.
  • A hand-built design system (tokens, type scale, spacing) so 15 locales feel like one product.
  • Full RTL support designed in from the start, not retrofitted.
  • Every drink gets a real photo — trust at a glance beats a prettier illustration.
Live on the App Store, fully offline, 15 countries — designed, built, and shipped by one person.
03

FireCode AI

Visit ↗
0→1 SaaS for the fire-alarm trade · React · Supabase · Claude

Fire-alarm designers juggle code books, spec sheets, and CAD plans across a dozen tools. I lived that world for six years — so I designed the product I wished my industry had.

Design decisions
  • Customer intuition from the trade itself: features map to real workflows (submittals, takeoffs, RFIs), not imagined ones.
  • An in-product AI assistant (Claude) designed as a colleague, not a chatbot bolt-on.
  • Dense-data screens with deliberate hierarchy — contractors scan, they don't read.
In production with a first paying customer — scoped, designed, built, and sold by me.
04

Interactive applications

Four-lens site ↗
Résumés as products · four-lens site, streaming chat, this site

Applications are the one product where the user (a reviewer) gives you 30 seconds. I treat each one as a consumer product with a single job: make the next click irresistible.

Design decisions
  • Each application gets its own designed artifact in the company's visual language — this site is the Kalshi edition.
  • Motion is choreographed to guide reading order — reveal, settle, then invite interaction.
  • Every claim links to a live, working thing. Show, don't tell.
A four-lens interactive résumé, a real-time streaming chat study (xai-resume.pages.dev), and the site you're reading.
Designed & built from scratch for this application — React · Next.js · TypeScript · Framer Motion. Source ↗claytonryanyoung@gmail.com