I built a PDF editor… and realized most of them work in a way people wouldn’t expect

I built a PDF editor… and realized most of them work in a way people wouldn’t expect I’ve been working on a PDF editor recently, and I went down a rabbit hole I didn’t expect. At first, I thought building a “simple” PDF tool would be straightforward: upload a file → edit → download. But the deeper I went, the more I realized something surprising: most online PDF tools don’t work the way users think they do. The assumption everyone makes When you upload a PDF to an online editor, most people assume: the file is processed instantly maybe some light transformations happen then it’s sent back In reality, a lot of tools: upload your file to a server store it (sometimes longer than expected) process it remotely sometimes pass it through multiple services Not necessarily malicious — but definitely not always transparent. The real technical challenge I tried to approach the problem differently. Instead of focusing on features first, I asked: what would a clean and predictable architecture for