Are you sure about adding the type hints to the description? I don't see it adding much, and the type hints are pretty long and messy.

But yeah, I can add bullets if you want

Thanks for the comment, I've gone in and implemented the changes you suggested. Can you check to see if it's better now?

Also, do you think the default error is too strict here?

yes, thank you for suggestions

But I mean the start index and end index not index that count from 0 as list.

Your index pairs seem to start from 1

... if so, the last index is inclusive too, which is not logical either.

Removed input validation. sequence length will always be equal to number of columns

I didn't think of:
test.expect_error("You must raise a ValueError", lambda: rename_columns(df, names), exception=ValueError)

It should be test.expect_error("You must raise a ValueError", lambda: rename_columns(df, names), exception=ValueError).

Related, but very different. One method of calculating the euler constant is completely different from another. This is a good, moderately difficult, Kata.

Looked good to me. No similarity between this Kata and the supposed similar Kata claimed in the thread below. You should publish.

related, maybe? https://www.codewars.com/kata/5a8d63930025e92f4c000086 (note: note implying anything here: I didn't do that one)

Added an example to the description. As of the link you provided, I don't want to add it. Firstly, because this is a well known math problem, and the users can find info about it easily even without any exact pointers (I got the wikipedia article as the first link by googling the name of the kata). Secondly, this article only explains the theory behind the problem, and includes lots of obscure math stuff, so I doubt anybody will benefit from reading it.

Random tests added