It was with an AI, asking some UX-related questions and assessing them based on it. The questions were simple but tricky. I feel it watching out for certain keyword usage in line with the questions asked, and if you don't say the right things, it grades you low