[pm90]

They aren’t measuring your ability to create value but in how you approach problems you don’t know the answer to. In many cases you will get a pass even if you don’t solve the problem, but can talk about what the solution would look like, what the interesting parts of the problem are, and so forth.

[aisengard]

But if you know the answer to all of them, what are they really measuring? Once you study enough there's not really much new under the sun. It's a really fatal flaw in the typical interview process.

[pimmen]

A good interviewer could ask some questions to find out if the answer is just memorized or if they understand the concept. And it would test your willingness to educate yourself on a domain, which is something I would value in my team.

And I say this as someone who doesn't particularly like the modern tech interview style, I think actually trying to build something is better. However, it does have some merit in measuring a candidate's ability even if it's not necessarily that great.