Amazon Development Center U.S. Software Engineer interview experience: grid paths and RAG

My on-campus Amazon SDE Intern process began with an online assessment. It combined coding with a Leadership Principles work simulation and a feedback or workstyle component. Only some students from the campus drive were shortlisted for the OA.

The coding included sorting IDs by frequency with smaller-ID tie-breaking, and an array-transformation minimum-cost problem based on greedy reasoning and observation. Round 1 was a one-hour Zoom interview with live coding. After a quick introduction and language choice, I worked through shortest path on a grid with obstacles, then follow-ups about BFS versus DFS, avoiding cycles, and complexity. A modified version asked for the minimum walls to break, with additional edge-case discussion.

Because I mentioned RAG and GenAI work, I was also asked about hallucination verification and semantic search across log files with millions of lines. We discussed chunking, embeddings, and vector databases for scaling. Round 2 was more behavioral and problem-solving focused, grounded in Leadership Principles scenarios. I was rejected, though I found it one of the most holistic assessments I had done: coding, DSA follow-ups, and GenAI/system thinking were intertwined, and I stayed engaged throughout.

Location: Hyderābād. Overall feedback: Neutral. Offer status: No offer.