It's TeX in wasm implemented by writing a pascal to wasm compiler, which seemed like a novel approach. (TeX is written in pascal and typically translated to C before building.)

TeX itself is a 500kb wasm file (compresses to 90k), but the images get much bigger with things like LaTeX pre-loaded.

(The author uses it to run tikz in a browser in this repository: https://github.com/kisonecat/tikzjax)

A few years later, when PDFs were more prevalent, I found out in an interview that the .pdf which I had sent in had been converted to plain text and then uploaded into a database of résumés and that it was essentially illegible due to issues with the encodings of the old-style figures, small caps, and ligatures which I had used.

Perhaps it would be helpful to ensure that the PDF is tagged so as to ensure that any automated processing results in a useful extraction of text?