> I have found pair-programming with a dev on a current problem you have in your codebase to be the best litmus test.
This has its drawbacks too. You're biasing towards familiarity with your code base's specific tech stack. Good engineers can learn new tech stacks in days/weeks, so that's not that important to screen for.
Plus "everyday" code is a lot of plumbing and busywork, so it's not as good a screen for intelligence. Algos at least cut to the "hard stuff".
> Like those CS grads in my class who... passed because they were good at memoizing.
Getting good at memoizing would be a perfectly fine reason to pass a dynamic programming class. :-)
This has its drawbacks too. You're biasing towards familiarity with your code base's specific tech stack. Good engineers can learn new tech stacks in days/weeks, so that's not that important to screen for.
Plus "everyday" code is a lot of plumbing and busywork, so it's not as good a screen for intelligence. Algos at least cut to the "hard stuff".
> Like those CS grads in my class who... passed because they were good at memoizing.
Getting good at memoizing would be a perfectly fine reason to pass a dynamic programming class. :-)