Comment by floatrock

> coming up with an efficient solution is a large part of the challenge

More of my work tends to be "rapidly adopting solution to additional and changing requirements" than "come up with most efficient solution", so why are we interviewing for something where in practice we just throw a couple extra instances at it? (Your specific job role may vary, of course, but I usually just increase the scaling factor)

Author's point is that coming up with the most efficient solution might not actually be a good measure of your real-world performance.

And that's been a longrunning critique of leetcode, of course. However, this is a neat framing where you can still use the same problems but give solutions that perform better when measured by "how adaptable is this to new requirements?"