The problems on this site were created using neural nets to automatically extract positions for each rank from high-level games where the neural net thought the next move would be instinctive for a pro but might be educational or non-obvious for players of that rank.
Trying to get back into the game a little, and I’ve just noticed that https://neuralnetgoproblems.com/ is still online! Whole-board positions from real games, asks you to predict the next move. It’s really good if you enjoy that sort of thing.
Thanks for the link! I never heard of that site but it sounds like an interesting approach.