Kata hint != kata suggestion. See https://docs.codewars.com/training/troubleshooting#post-discourse.

Also, this is a 5 kyu kata. Users should know how to efficiently compute fibonacci numbers at this level, and if they don't, they should be able to research how to by themselves.

fixed. your solution fails both the random and fixed tests consistently now

your solution writes past the end of the output buffer: you are printing a trailing '.', and then erase it. this is not robust: the worst-case output size is 16 i.e. strlen("100.100.100.100") + 1, so the caller of the function can expect that 16 bytes should be enough, but your solution needs more bytes than that. as a result of writing out-of-bounds, you corrupt the tests suite

it is your code that is wrong: you are adding trailing zeroes instead of leading zeroes to the binary string. please ask a question in the future instead of raising an issue ;-)

no, the problem comes from your code. you are assuming that the binary string you create is always 32 chars long, this is not true.

Please ask a question (not an issue) to ask for help in the future ;-)

Is it possible to implement this in Kotlin? When attempting a logical AND operation, garbage is returned. When I look at the code that compiles in Java, there is a completely wrong idea.
UPD:Implemented via string. But the test fails. Error with ip: 89.118.64.172.
In 10th representation it is 1500922028.
In the 2nd representation - 10110010111011001000000101011000. The first 8 bits 10110010 = 178. Am I missing something or is there an error in the tests?

If you were able to write a “slow” version, look at the output of numbers and the operation of the algorithm, look for a pattern.