My Amazon SDE intern interview process began with an OA (Online Assessment) that included debugging, data structures, and logical reasoning.

After clearing it, I had a 45-minute virtual technical interview.

The interviewer was friendly and started with a quick introduction, followed by one DSA question.

I was asked a BFS-based problem similar to Rotten Oranges (multi-source BFS). I explained my approach, edge cases, and optimized the code. Then we discussed time and space complexity.

Finally, I was asked a recent problem: Find the number of unique paths in a grid with obstacles.

The interview concluded with a few behavioral questions.

One key thing they asked me was:

"Can you optimize your solution further or reduce the space complexity?"

This came right after I solved the BFS-based Rotten Oranges-like problem. I initially used a queue and a visited matrix, and the interviewer pushed me to think if I could avoid the extra space.