I totally agree with asking the basic question and then adding scale and complexity. I don't know if I agree that using a trie at the outset reflects poorly on

I totally agree with asking the basic question and then adding scale and complexity. I don't know if I agree that using a trie at the outset reflects poorly on a candidate. They might see where you are going with the question, even though "you didn't mention scale." Good candidates are going to think about scale at least a little bit. That's not necessarily "lacking practical experience." It's not like basic tries or linked lists are super complex "fancy" data structures. I think the key is for both parties to communicate thought process. If you're concerned about overengineering then prompt them to explain why they didn't choose an array.