It may appear that way on the surface, but that's because it follows a typical pseudo-scientific formula of citing some a few specific results as evidence to give the argument an air of credibility, and then drawing broad and profound conclusions that are at best vaguely related to the evidence. He basically went from "there are biological differences between men and women that affect their preferences" (okay, I can maybe buy that) straight to "women may not be as good a fit as men for these roles" and "Google should consider ending all of its diversity efforts". The evidence he cited (some of which is research that's been debunked) doesn't even come CLOSE to justifying his conclusions.

The reason there's such an uproar within Google is that the place is full of people with extensive training in mathematics and logic, who can easily spot when you submit a "proof" that is missing about 10,000 steps between the low level lemmas and the top level claims. Combine this with the fact that the claims are questioning the right of 1/3 of the company to exist there, and of course the reaction is going to be incredibly visceral. I honestly don't know how anyone could read more than a paragraph of that document without their bullshit detector exploding.

I suspect the pretend-polite tone makes it even worse. Women are tired of hearing these prejudices, but they're often abrupt, poorly spelled missives that are easy to ignore.

When somebody wraps it into a careful sandwich of "But have you considered testosterone? Just asking..." for 10 pages long, it feels like the author is seriously working hard to undermine your professional status, not just spouting on a drunk night.

Ultra mild sexism is great, but still sexism.