I would most definitely discourage this advice, in my opinion it's dangerous. Instead, I would focus on slowly and "casually" refactoring bad parts (without putting significant extra hours at all), while making sure that the code is actually badly written, and it's not just "not written your way" or "code I can't understand because I didn't study it enough". I have seen many engineers being victims of these two points. Sometimes the code just appears badly written because you don't have enough context to see that it's not.