Group credit card customers with no labels, pick K and describe each group for the business.
New to this? See the worked example firstK-Means finds groups in data that has no answer column, such as kinds of credit card customers. In these questions you scale the columns, choose K with the elbow and silhouette score, and describe each group in real units.
Interviews ask how to choose K, why scaling matters and why different starts give different answers; every one of these is covered here.
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.