Your question is Flexible Multi-Class Classifier. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
Describe a time you built a model or algorithm to classify 10 classes with the expectation that the number of classes may change in the future, and explain how you kept the architecture flexible.
Explain the data representation, model choice, label-management strategy, training procedure, and evaluation approach. Discuss how you would add a class, prevent catastrophic degradation of existing classes, and distinguish genuinely unseen classes from known classes. Include production considerations such as model versioning, backward compatibility, retraining, and monitoring.