I mean it's also a pretty good filter by age: if you're middle aged, working and have children, then you're not going to really have the time to get through memorising a bunch of leetcode.

Which is handy because Google et al don't really want employees who look at the in-office perks and wonder why they'd be appealing at all if you're just going to go home at the end of the day.

If you're a good middle aged engineer, you should have enough former coworkers happy to hire you that you don't need get on the leetcode treadmill.

This is such a funny thing to read, because I see new L6 and 7 tech managers hired (above me) all the time. They're often middle aged people with kids. They don't bemoan this stuff (and yes, while they have fewer of them, they still need to solve a whiteboard problem or two to get hired)

But you're seeing the outcome of a selection process. You might feel differently if you saw the top of the funnel.

In what way? If you mean that there might be management wizards who are just fantastic but we skip, perhaps. But that isn't the point I was making, just that despite what people think, these interviews don't actually filter out experienced people with families and kids. Those kinds of people get hired regularly.

Well you don't actually know that, because you don't know what the top of the funnel looks like.

Consider a world where every L6+ manager candidate has a family and kids. However, only 5% of these can make time to do the LC stuff (revising most likely if they've been managing for a while).

So your company will end up filtering out 95% of candidates immediately, but people in the company will say that there's no problem as all of the successful candidates have a family and kids, so it can't be biased.

Don't get me wrong, I'm a fan of technical screening for managers, but I'm making the point that you can't look at the output of a funnel and conclude that everything's fine, you need to look at the complete funnel to understand this.

> The flip side is the candidate that memorized the book, prepared like it's a test at school, and is useless on the job, can't operate at all outside that memorized domain, can't write the simplest bit of real world code.

I think this leaves out the possibility of high mastery. I think it is instructive think about algorithm skills as similar to those for algebra. If the most advanced algebra you ever solved was linear equations, remembering how to solve a linear equation may be difficult. However if you're skilled at solving advanced calculus problems by hand, your understanding of algebra is so deep that you could re-derive the process as necessary. Many of us with a lot of math experience will never forget how to solve linear or quadratic equations barring steep cognitive decline and this is also so true for certain classes of algorithms. Personally, I've studied graphs and graph algorithms beyond what you might find in just CLRS or other introductory algorithms textbooks and as a result I can re-derive many classic graph algorithms on the fly, no memorization necessary. I have done this interviews as have people I know.