2 ms·
Fine, but don't you think companies should have some filter to choose candidates? In my company I interview often, and most candidates don't know how to code. A
by gnulinux 8y ago
Fine, but don't you think companies should have some filter to choose candidates? In my company I interview often, and most candidates don't know how to code. Asking CS hw questions is a signal that predicts if they can code or not. If you're not a CS major it's ok but in general if you can't figure this question in 15 mins I don't want you to push code to my codebase. Some level of competence is ok. And I don't even work in a FAANG this is a random no name startup in Boston.
- hinkley 8y agoThere's one particular category of question that I've started building up a question set in, in part because the last few coworkers who weren't nearly as useful as you would hope all struggled with it: Data transformation. Most of us are at least partly in the business of data transformation. Few of us should be implementing trees (you want coworkers who are informed consumers of data structures, not people who think reimplementing solved problems at work is a reasonable or even sane thing to do). The simplest one I have is this: Hand them a data structure containing a list of lists, have them generate a list of the grandchildren ordered by some property of the grandchild. Like take a list of cars or computers grouped by make and model and give me all of the models sorted by price, or by popularity. If that's too simple, filter by another criteria and do a little data cleanup. I've seen people drop elements, get the order wrong, scope variables too broadly and confuse themselves into grabbing data from a previous iteration, or forget to grab it entirely. And this is exactly the sort of thing that puts your team on the spot for having dumb bugs in production. Not inverting binary trees, or fizzbuzz.