Apple Software Engineer Interview Questions

I would first sort the intervals by their start time. Then I would scan through them and either merge the current interval with the next one if they overlap, or move on to a new interval if they do not. The reason I choose this approach is that once the intervals are sorted, the merge decision becomes very easy. The time complexity is O(n log n) because of sorting, and the scan itself is O(n). I would also mention edge cases like touching intervals, nested intervals, and already sorted input.

Before designing it, I would clarify whether this is only local device storage or a cloud-synced photo library with sharing and search. At a high level, I would separate metadata storage from actual image storage, because users need quick browsing even when the image files themselves are large. I would also think about thumbnails, indexing, and sync so the app feels fast when users scroll through many photos. The trade-off is between simplicity and richer features like shared albums, face grouping, or advanced search. I would also mention privacy and access control because photo apps handle very sensitive user data.

I would first clarify the backup scope: which file types are included, how often sync happens, and what level of end-to-end protection is required. My default design would encrypt data before upload, store metadata separately, and make sure only the user can decrypt the backup content. I would also think about incremental backups so the system only transfers changed data, which keeps the experience fast and reduces cost. The reason I choose this architecture is that privacy and reliability both matter, especially for personal data. I would also discuss key management, recovery flows, and what happens if a device is lost.

I would first think about sorting the whole array, but that is more work than necessary. To optimize it, I would use a min-heap of size k so I only keep the largest k values seen so far. If I want an even more efficient average-time solution, I could also mention quickselect. The reason I choose the heap first is that it is easier to explain and safer to implement during an interview. The time complexity is O(n log k) with the heap, and the space complexity is O(k).

I would use a sliding window with a hash map or set. As I move the right pointer across the string, I would shrink the left side whenever I see a repeated character, so the window always stays valid. The reason I choose this approach is that each character is processed a limited number of times, which gives O(n) time. The space complexity is O(min(n, alphabet size)). I would also mention edge cases like an empty string and strings with all unique characters.