The issues being brought up are very similar to the things talked about in PG's "The Lesson to Unlearn" [1], i.e. "hackable" tests in university. IMO, I don't

The issues being brought up are very similar to the things talked about in PG's "The Lesson to Unlearn" [1], i.e. "hackable" tests in university. IMO, I don't think there is anything wrong with expecting candidates to be knowledgable about basic algo/data structure for a SWE role -- instead, the issues are with the evaluation process. Namely that it's become hackable using these preperation services (resulting in false-positives), it doesn't capture or indicate a candidate's real-world knowledge/experience/abilities, and the typical questions and whiteboard approach is likely outdated when there are probably better alternatives. [1] http://www.paulgraham.com/lesson.html http://www.paulgraham.com/lesson.html