Not really. As somebody who has seen both sides of a hiring process like that, the reality is the vast majority of candidates are just awful. You can probably rule out nearly 75% of applicants right off the bat. Add in hiring laws, bureaucracy, immigration, negotiation, candidates withdrawing, and finding a good person who works for your team can take months.
I was under the impression that it was sort of like Harvard: lots of applications, but a good chunk of those denied could have just have easily been accepted if there was space
You seem to have experience with their hiring and I have none (only applying) so I have much more faith in what you're saying than my preconceived ideas.
As in, a significant chunk of candidates cannot write code, have outright lies on their CV, have never used anything other than visual basic, etc...
In addition, basic statistics suggests you need all new hires to be better than 50% of the existing company, or your average skill level will go down, which makes hiring just hard mathematically.
> Can't you rule a lot of those candidate (other than CV liars) out algorithmically?
I think all companies do this, but it's really easy to game the metrics. People do it unconsciously, even.
> This seems especially difficult when hiring graduates/junior roles due to the lack of unknowns.
Yep, so it takes hours of interviewing per candidate, across dozens of candidates just to hire to minimise the risk for one role. At a global level, FAANGs are hiring loads, but at a team/local level there's never enough people due to specific team needs or turnover.
Cannot write code is very different from cannot write code on a whiteboard to solve a difficult problem in a limited amount of time in a stressful setting where you constantly have to keep talking and explaining what you are doing and cannot pause to think.
Yes, could you elaborate on the awful nature of most applicants? Do you mean that they are not fit to make software or that they are not Google material?
The former. Everybody knows top quality engineers are very hard to find, but there is also an inverse problem - the amount of programmers who know nothing but React.js or other <buzzword> is shockingly high.