Here's an easy ad-hoc Prolog program for the first problem:

    % Given a set of coin denominations,
    % find the minimum number of coins
    % required to make change.
    % IE for USA coinage and 37 cents,
    % the minimum number is four
    % (quarter, dime, 2 pennies).
    num(0). num(1). num(2).
    num(3). num(4). num(5).
    ?- num(Q), num(D), num(P),
       37 is Q * 25 + D * 10 + P

(Posting again what I already posted two days ago [2] here)

[1]: https://quantumprolog.sgml.net/browser-demo/browser-demo.htm...

[2]: https://news.ycombinator.com/item?id=45205030

Of course, the challenge is that the next question after solving a leetcode problem is often to explain and optimize the performance characteristics, which in prolog can get stupidly hairy.

I've actually used pseudo-prolog to explain how to solve leetcode problems to a friend. Write the facts, then write the constraints, and then state your problem. Close to the last part, they've already understood how to solve it, or at least how to write the program that can answer the question.