CD Plates
Vibe coded an entire iOS app that lets you instantly identify Swiss diplomatic license plates.
This started as a curiosity project in Geneva. Swiss diplomatic licence plates follow a numbered system that identifies the country or international organisation behind the car. I kept seeing these plates around the city and wanted a faster way to know who they belonged to.
The information technically existed, but it was scattered, incomplete, and buried across different sources. Some details were hidden deep in official pages, some were easier to find in PDFs, and the experience was not designed for someone quickly trying to identify a plate on the street. I wanted to turn that scattered information into something instant, searchable, and useful in the moment.
The result was CD Plates, a lightweight iOS app for looking up Swiss diplomatic plate numbers and identifying the related country, organisation, or mission. The app combines local CSV data for fast offline lookup with Supabase-powered data where live updates are useful. I also collected and organised flag assets, and scraped the UN Blue Book to add richer mission-level details.
A big part of the product was the speed of the experience. From my own use case, if you see a moving car, you may only remember the number for a few seconds. So the app is designed to open directly into search, with the native keyboard ready immediately. The goal was to remove as much friction as possible: open the app, type the number, get the result.
This was my first serious AI-assisted iOS project, built before tools like Claude Code and Codex were released. As a beginner coder, I used a manual AI-assisted workflow: asking ChatGPT to explain concepts, suggest code, help debug errors, and clarify SwiftUI patterns, then copying code into VS Code and Cursor, testing it, fixing what broke, and learning how the app worked step by step.
After launching, I shared the app through Reddit and LinkedIn. User feedback from social posts helped shape later improvements, including added statistics, spotted tags, and richer details around diplomatic missions. The app now sees regular local usage, with thousands of lookups every month.
What made the project valuable was not only the app itself, but the full process: noticing a real local problem, gathering messy data, building a fast mobile experience, learning iOS development through AI assistance, shipping to the App Store, and improving the product based on real users.
