I applied through a recruiter.

The first phone screen with a recruiter asked me to describe API building experiences. The recruiter shared the company culture of strong ownership attitudes. Then I was sent a link to an online assessment. The assessment had four levels, and the expectation was to complete all four levels during the allotted time of 90 minutes.

You are asked to design and implement an in-memory cloud storage system that maintains a mapping from file names to their metadata (size) and supports both single-user and multi-user operations.

Assumptions to make explicit: File names are unique per user. The early single-user functions operate in a special default user namespace. Sizes are non-negative integers (bytes). Capacity is the maximum total size (sum of file sizes) allowed per user.

For get_n_largest(prefix, n), return the n largest files across all users whose names start with prefix. Each result should identify the user.

merge_user(u1, u2) combines user2 into user1; user1 remains, user2 is deleted. Capacity becomes the sum of both users' capacities. If file name collisions occur during merge, rename the incoming conflicting file(s) from user2 by appending a suffix to ensure uniqueness (e.g., "name (merged 2)").

backup_user and restore_user operate per user and do not affect other users' files or capacities.