Recommend a workout programme from similar members, with scaling, distance and the right K.
New to this? See the worked example firstK-Nearest Neighbours suggests an answer from the most similar rows, such as the workout programme of members like you. In these questions you measure distances, see why scaling matters, and choose K with cross-validation.
It is a common interview topic because it shows whether you understand distance, scaling and overfitting with a very small K.
If you are new to it, read the worked example first, then solve the questions in order: the early ones explore the data and train a first model, the middle ones test it on rows it has not seen, and the last ones tune it and turn the result into a business decision.