I'd say asking a candidate to implement levenshtein distance in an interview (probably not the best question to ask I'd say) is more about creating a level playing field for candidates who might not have "attended and spoken at relevant conferences" or have "a bunch of apps doing mission-critical work in production", but might be thoroughly productive and successful at XCorp nonetheless.
In fact, a candidate might have experience in something completely different but maybe just as impressive - publishing a paper on an optimization for an algorithm, creating a programming language, submitting Linux kernel patches, etc. How would one differentiate a candidate like that with someone with "a list of public repositories on GitHub as long as your arm"?
In fact, that's kinda the point. Each candidate should be evaluated on the unique experience they bring to the table. Trying to create the illusion of a level playing field by standardizing on generic code challenges in interview exercises is just that, an illusion.
In fact, a candidate might have experience in something completely different but maybe just as impressive - publishing a paper on an optimization for an algorithm, creating a programming language, submitting Linux kernel patches, etc. How would one differentiate a candidate like that with someone with "a list of public repositories on GitHub as long as your arm"?