But let's nitpick a bit. I find it interesting that the arguments made in the article are about the intention of the programmer.
Perhaps an explicit nil does signal to me that the nil is intentional. But does that mean that all implicit nils are unimportant? For a skilled programmer like the OP, it probably does mean that! But how often do I actually know who wrote the code I'm reading? Of course I can git blame, but maybe the author left the org before I joined. What kind of programmer were they? Are implicit nils important or unimportant to them? Perhaps it's easy to argue that they should be unimportant. But in reality, I don't know what the author of a particular method was thinking while they were writing it. In fact even an explicit nil might be unimportant. Maybe somebody replaced a `raise SomethingError` line with `nil` to hot-fix an urgent bug.
I guess from my perspective as a long-time Ruby developer, all nils are important. It doesn't matter what the intention of the programmer was. Ruby is very expressive and very dynamic - anything can return nil and there are a hundred ways to do it. If your nil is intentional, write a test case for it. If you have no test suite in a Ruby project, then may God have mercy on your soul.