NameError using multiprocess pool with function which calls another function

I would like to use multiprocessing in my jupyter notebook and AFAIK, the multiprocess package is the way to to that. My function does quite a lot of stuff, so it calls multiple subfunctions, but this throws a NameError for the name of the subfunction, unless the subfunctions are defined in the function beein called.

import multiprocess as mp

def test_int():
    return 1

#This throws import multiprocess as mp
def test_int():
    return 1

#This throws "NameError: name 'test_int' is not defined"
def test():
    return test_int()

#This works
def test2():
    def test2_int():
        return 2
    return test2_int()

if __name__ == '__main__':  
    with mp.Pool(20) as pool:
        for _ in range(20):
            res = pool.apply_async(test, [])
            print(res.get())

As the functions are defined before the process is started and even before the called function is defined, I don't see a reason why this wouldn't work, but it doesn't.

The second varient works for now, but gets messy fast with more and bigger functions, so is there a way to make this cleaner?