You are arguing that. He doesn't.
What I see argued there is the tendency for women to prefer something like early childhood education or psychology over software engineering.
It has nothing to do with suitability at all.
See also: Why Brilliant Girls Prefer non STEM Subjects[1]
The studies it cites show that women who are good at math (SATs, I think) also have good verbal skills, so they simply have more options. And apparently people of either gender who have both options open to them prefer non-STEM.
That is even an ability difference (whether biological or not is irrelevant, though SAT scores appear to have a strong genetic component) that has nothing to do with women being "less suitable" for STEM and yet fewer women in STEM.
So your logic that any type of argument, even if it's just preferences, has to automatically mean that there is a claim (if hidden) women are less suited is quite simply wrong.
Apart from the mistake pointed out elsewhere of taking a statistical argument about populations that tries to explain population differences (and not individual differences) and turning it into exactly the essentialist argument you find objectionable.
[1] https://www.psychologytoday.com/blog/rabble-rouser/201707/wh...