Selected work
Built to ship and built to last
Websites I designed and built for clients, brand included, then handed over. And my own apps, built with Flutter and Firebase.
- Client site, end to end
- 01
- Shipped apps
- 04
- Downloads
- 100k+
Client work
Client websites, end to end
Consultants, coaches, advisors: one person whose reputation is the product. I design and build the whole thing: identity and logo through to a running site with a CMS, a blog and a newsletter behind it, handed over so you can keep changing it without me.
Brand & corporate design
Logo, palette, type and motion: the identity before the site.
A site you own
Pages, blog and newsletter in a CMS with your name on it.
Guardrails, not handcuffs
Security, speed and correctness, checked automatically.
Handover that survives me
A codebase where you can change copy and colour without me.
In practice
von Wartenberg Leadership
The complete web presence for Dr. Peter von Wartenberg, who advises executives and organisations on leadership. Brand first: the “Rising Arcs” mark and the claim “Learn to Serve & Protect”, then a bilingual site built to carry them: Next.js and Tailwind, Payload as the CMS on Neon Postgres, Brevo behind the newsletter, Vercel for delivery. Pages, blog and mailing list belong to the client, and the release pipeline keeps a cosmetic change from breaking anything underneath it.
Scope
- Brand identity
- Logo: the “Rising Arcs” mark
- Corporate design
- Motion & animation
- Animated logo & level states
- Web design
- Payload CMS setup
- Newsletter & blog
- Bilingual DE / EN
- CI/CD and guardrails
Stack
- Next.js
- Tailwind CSS
- Payload CMS
- Neon Postgres
- Brevo
- Vercel
- TypeScript
System
Machine learning, in production
A model in a notebook helps nobody. My day job is the platform that gets models into production at Sixt and keeps them there: monitored, governed and able to grow.
ML platform architecture
the training, deployment and monitoring path a team uses daily, not a single pipeline that works once
Prototype to production
taking a model that runs on a laptop and making it something an on-call engineer can be woken up for
AI features inside a product that already exists, scoped so the useful part ships before the budget is gone
Direction for a team already building it
architecture, tooling, and the handful of decisions that are expensive to reverse
Projects
Also shipped
MuscleMate - Gym Buddy
A workout log for the gym: exercises, sets, goals and progress over time. Built in Flutter, with Firebase keeping it in sync.
No longer published
MoneyMate - Expense Tracking
An expense tracker that sorts spending into categories and shows, per category, how you are doing against your budget.
No longer published
SpendWise - Mindful Spending
A spending app that asks one question before you buy: do you actually want this? Prompts and habit tracking instead of after-the-fact budgets.
No longer published
Want something built like this?
Tell me what you need and roughly when. I'll be straight with you about whether I'm the right person for it.