Eventfrog Entry App
Ticket Scanning That Cannot Fail
Three months before a 25,000-ticket event it was clear: the old entry app would not survive the load. I designed and built the new one. It validates tickets without an internet connection, and entry ran without a single incident.
Role | Lead Client-Side Developer, Technical Project Lead |
Duration | 3 Months (concept to deployment) |
Tech Stack | Angular, Ionic/Capacitor, TypeScript, IndexedDB, Web Workers, WebSockets |
The Starting Point:
25,000 Tickets and Three Months
This project was born from pressure. A major event with 25,000 tickets was three months away, and the existing entry app would fail under that load. A new solution had to be built from scratch, and it had to work on its first day.
Two engineers had three months for frontend and backend. As the sole frontend developer, I was responsible for the client application: more than 25,000 lines of code that could not fail under full load.
My Approach:
Offline First
The architecture followed one principle: the app always works, even without internet.
The foundation is a robust offline mode. All data for the selected event and its tickets lives in an IndexedDB on the device. The critical logic, checking validity and preventing duplicate scans, therefore runs directly on the device without a server. Heavy processing runs in Web Workers so the UI stays fluid.
On top of that, I built the synchronisation. As soon as there is a connection, the devices exchange data with the server and with each other over WebSockets. When one device scans a ticket, every other device at the same entrance knows immediately. Using a ticket twice is impossible. If the connection drops, a resume mechanism with a tamper-proof timestamp catches up later. A REST endpoint serves as the final fallback.
Real-time statistics and a ticket list with live status complete the app.
The Result:
Not a Single Incident
We hit the deadline. The new Entry App was used at the 25,000-ticket event, and entry ran without a hitch. The app handled the scan volume reliably.
It has been in service for Eventfrog every weekend since.
What this project shows about how I work:
- I build offline-first applications under time pressure.
- I implement real-time synchronisation with WebSockets.
- I move heavy computation into Web Workers to keep the UI responsive.
- I deliver software that holds up under full load.