Python can be fast, but think of it this way: you can have 90% of your code base written in Python and the rest 10% written in C or C++, then wrapped around with Cython. This will be wayyyyyy faster than writing 100% of your code in C or C++, and from an industrial perspective, this is much much better than anything that other languages can make.
Python is slow as balls, but it’s still extremely versatile and if you can’t make an application fast enough or meet your requirements given enough time without having to change your entire code base, then you have failed as a programmer. Python isn’t the best tool for everything, but that’s not the point of the programmer.
In data structures and algorithms, you don’t find something that fixes all problems, you find one that fixes your problem. It’s just that more often than not, it’s Python that fix your problem time and time again