[jasonpeacock]

No, you can't make those assumptions.

People graduate from high school who are functionally illiterate. Similarly, people graduate from CS programs who are functionally incompetent.

I've interviewed many people who can talk the talk, have an impressive resume, can talk details about their projects... But fail a simple coding question. My go-to phone screen question is to find the common elements from two lists of integers, and half the candidates can't do it in better than O(n^2). I'm not even looking for the perfect solution, just something that is better than brute-force.

It's the "do original work" that is the hardest for many people. They can modify existing systems, plumb together interfaces, etc. without a problem, but given a fresh problem and a blank sheet of paper they are stuck.

I've found the same people are also weak in their fundamentals...

[zerr]

The "original work" vs "plumbing things" distinction is very vague - we all stand on the shoulders of giants. E.g. PageRank also can be seen as Algebra stuff plumbed together.

Even if we go with your definition - why do you assume that someone who practiced leetcode/hackerrank stuff and thus are able to pass interviews is good with the "original work"?

I tend to think that the current interview practices have two (eventual) goals:

1. Hard to change jobs because practice is needed.

2. Tests how dedicated/desperate the candidate is - willing to allocate time for preps.