I hear that accusation a lot, but I’ve never really seen a company do that. Having been in the other side of the equation many times, the effort required to review code that’s submitted, especially when you don’t know the author is probably not worth it.
The case I detailed above was a fairly isolated subsystem that didn’t really require any knowledge of their system to work on. They probably chose it because A. It was a readily available problem with an existing solution that you could reasonably expect to be solved within a couple of hours , and B. the CTO thought he had a cool solution.