Amazon

Amazon SDE Interview Experience I recently interviewed for an Amazon SDE role. The process had two rounds, with a mix of DSA, projects, Leadership Principles, and Generative AI. My result was pending at the time of writing this. Round 1 - DSA + Projects + Leadership Principles The round started with DSA. Weighted Non-Overlapping Intervals Given arrays of start times, end times, and weights, the task was to find the maximum total weight of non-overlapping intervals. I first explained the brute-force recursive approach and then optimized it using Dynamic Programming + Binary Search. After sorting intervals by end time, for each interval I found the latest non-overlapping interval and used: `dp[i] = max(dp[i-1], weight[i] + dp[last_non_overlapping])` The interviewer asked me to clearly explain the approach and dry-run it. Equal Sum Partition The second question was to determine whether an array could be partitioned into two subsets with equal sum. I explained the standard Subset Sum DP approach. The follow-up was interesting: n ≤ 20, but the values can be very large. Since sum-based DP would be expensive, I suggested Meet-in-the-Middle, generating subset sums for the two halves and checking whether they can form the required target. After that, due to limited time, we moved to Amazon Leadership Principle-based questions mostly scenario-based and related to my projects. I was asked about situations I had faced, how I handled them, and the decisions I made. Being comfortable with my projects helped a lot here. Round 2 Gen AI + Coding The second round focused more on **Generative AI and practical engineering scenarios. I was asked: Where have you used Gen AI? When did it work well? When did it fail? How do you verify AI-generated solutions? How do you decide whether to trust an AI-generated answer? There was also a coding problem involving m employees, n chairs, 5 working days, and each employee needing to work from the office for 3 days. The interviewer first asked me to derive the feasibility condition and then construct an assignment if possible. The key condition I identified was based on total weekly capacity: `3m <= 5n` Then the discussion moved toward constructing a valid assignment while respecting the daily chair limit. Overall Experience The interviews were quite interactive. The interviewers focused not only on getting the correct answer but also on **how I approached the problem, explained trade-offs, handled follow-ups, and communicated my reasoning**. My biggest takeaway: for Amazon, prepare **DSA + projects + Leadership Principles together**. During DSA, always start with brute force, optimize based on constraints, explain complexity, and dry-run your solution. Also, know your strongest projects in depth because they can become a major part of the interview.