I built a small event-driven video transcoding pipeline to explore how large-scale media systems behave.
The live site lets you upload a video and use the pipeline in action. The architecture, tradeoffs, and reasoning are documented in a blog post, and the code is open source.
Blog: https://v0id-user.mataroa.blog/blog/i-tried-to-understand-ho...
Repo: https://github.com/v0id-user/tcoder
Happy to discuss the design decisions, limitations, and what I’d do differently :)