Yes, the majority of the app is written in TypeScript. It uses github.com/lovell/sharp. I think there are a lot of great tools in JavaScript land for computer vision, but maybe not as many code examples or theory explainers that go beyond resizing images for web servers. Some features are missing here and there, but the pieces written in other languages to fill the gap are a few and don’t need as much maintenance.
Overall, JavaScript seems like an ideal solution for this project. The worker, server, and client all use the same language, which is very convenient and easy to read. At least for someone so used to writing everything in JavaScript 😅
Dmitri Mar 25, ‘25 edit