My alma mater Harvey Mudd College has rough gender parity in Computer Science so it's clearly achievable for colleges that make a focused effort. https://www.hmc.edu/about-hmc/2018/05/15/harvey-mudd-graduat...
Harvey Mudd is a very elite college with considerable resources, and it has a relatively small class, especially compared to large public universities.
https://www.hmc.edu/institutional-research/institutional-sta...
It looks like there were 27 CS majors out of Harvey Mudd in 2017. We can probably add in another 20 from the joint CS-Math degree. So, about 50.
Here are the numbers for Berkeley:
https://eecs.berkeley.edu/about/by-the-numbers
Berkeley graduates almost a thousand students in EE or CS every year. The accounting is tough here because the degrees at Harvey Mudd and Berkeley are different, so I may be over or under counting, but I hope you can see my point - Harvey Mudd succeeds in a way that might not apply once you try to reach scale.
Consider that 27% of the high school students who take the AP Computer science test are women. This next part is a big assumption, but let's say that this reflects the percentage of students in the current pipeline, and that a "5" on this test clearly represents a high talent student. It's not the only way to identify such as student, but it is a strong indicator.
Harvey Mudd could probably fill it's entire class, all 50 spots, with women who got a 5 on the AP CS test. Harvey Mudd could probably go up to 100% women without meaningfully compromising admissions standards. Remember, this debate is about ratios. The author of the controversial piece is not under fire for saying that talented women don't go into CS, he's under fire for saying that 20% is a natural equilibrium.
Yes, Harvey Mudd could fill its small but very elite CS class entirely from the 20% of the top cohort. But could Berkeley, and Michigan, and UW, and Georgia Tech, and...
I'm not saying that these schools can't achieve parity, nor am I arguing that they shouldn't. But the process for getting to parity across lots and lots of large schools is very different from what Harvey Mudd can do. There would still be lots of men with "5"'s looking for a spot, but not as many women - simply because there aren't as many women taking the test.
At some point, the elite programs will be unbalanced if they simply reflect the pipeline. That doesn't mean they shouldn't try for more equity, but we should recognize that Harvey Mudd's approach won't necessarily scale, it may reflect the recruiting advantages of an elite, well funded school with lots of resources and very few undergraduates.