I recently interviewed for a software engineering role and progressed until the second round.

The first round began with some behavioral questions, such as:

• Tell me about yourself
• How do you approach a problem you’re solving?
• How do you handle something unfamiliar?

After that, I was given two coding problems. The first one was a LeetCode medium-level problem (also tagged on LeetCode), which I was able to fully solve, including dry runs and edge case discussions.

The second problem was similar in concept to the Alien Dictionary problem, but not the same. I explained both the brute-force and the optimal approach, though I didn’t get time to fully implement it.

The second round was more focused on algorithms and deeper problem-solving.

I was asked a challenging question related to topological sorting, which wasn't from LeetCode and had a custom constraint twist. The interviewer joined about 20 minutes late but was kind enough to extend the session to give me the full time.

In this round, there were also a couple of behavioral questions, including:

• What do you want to be when you grow up?
• Why Bloomberg?

You are given a database that describes N six-sided boxes. One side of each box is open (no flaps). Each box has a unique integer ID and is defined by three dimensions: width, height, and length.

Your task is to determine which boxes, if any, can fit inside other boxes to consolidate space.

A box A can fit inside box B if its base (length × width) can be completely placed within B’s base, in any rotational order. This means both dimensions of A’s base must be strictly smaller than the corresponding dimensions of B’s base, even after considering rotation (i.e., min(LA, WA) < min(LB, WB) and max(LA, WA) < max(LB, WB)). Additionally, box B’s height must be greater than or equal to box A’s height.

You are expected to output valid box containment chains, where each box fits inside the next. An example of such an output chain might look like: A → C → Q → B or M → D → Z or R → O → E → Y, and so on. The idea is to show the maximum possible nesting or consolidation paths, based on the given dimension rules.