What happens the moment you drop a file in
When you drag a PDF onto the page, it's read directly into the browser's memory via the File API — the same mechanism any web page uses to let you attach a file, except here the file never gets sent onward in a network request. The WebAssembly engine then processes that in-memory copy and hands back a new file for download.
The entire round trip — read, convert, download — happens without the page making a single request that carries your document's content, which is straightforward to confirm in DevTools if you want to see it for yourself.