Amazon SDE-1 Full Loop Experience

I finally got the chance to go through the Amazon SDE-1 New Grad loop for AWS. There were four rounds, and each one felt quite different, heavily focused on coding and design, while others were more about my past experiences and how I handled different situations.
Role: SDE-1 New Grad, AWS
Experience: New Graduate, Fresher
Rounds: 4
Result: Offer received

Round-by-Round Experience

Round 1: DSA and Leadership Principles

I first discussed a situation where I faced a blocker while working on something important, how I handled it and what I would change if I faced it again.

For DSA, I received Path Sum II. I had to find all root-to-leaf paths in a binary tree whose sum matched a given target.

I solved it using DFS and backtracking, keeping track of my current path and remaining target.

After solving the standard version, I received a harder variation. I couldn't identify the complete approach immediately, but after a couple of hints, I was able to work through it.
Similar Problem:

Round 2: Low-Level Design

My behavioral discussion here was around a time when I took ownership of something outside my original responsibilities.

For LLD, I had to Design a billing service for a single AWS customer.

Given a customer's usage across AWS services and their respective rates, I needed to calculate the final bill, I modeled concepts such as customer, service usage, pricing and billing, and then explained how I would connect them.

As the requirements changed, I had to adjust my design and justify the classes and abstractions I had chosen.

Round 3: GenAI, DSA and Hiring Manager Discussion

Started this round with a discussion around GenAI and its practical applications before moving to coding.

My first problem was First Non-Repeating Character.

Similar Problem:

The second problem was a variation of Maximum Product Subarray. Along with finding the maximum product, also had to return the start and end indices of that subarray.

Because negative values could turn a minimum product into a maximum, I maintained both the maximum and minimum product ending at each position. I also tracked their starting indices so I could return the correct range.

I then handled follow-ups involving zeros, multiple negative numbers and single-element arrays.

The rest of my discussion focused on decisions and situations from my previous experiences.

Round 4: Bar Raiser

My Bar Raiser was entirely behavioral, with no coding question.

I discussed situations involving ownership, disagreements, failures, difficult decisions and lessons I had learned.

Most of the depth came from follow-ups. I had to explain why I made a particular decision, alternatives I considered, what my personal contribution was, the impact I created.

I made sure my answers focused on my own actions and decisions, rather than speaking generally about what my team had done.

Result

After completing the full loop and waiting for the final update, I received the Amazon SDE-1 New Grad offer for AWS in Seattle.

Preparing DSA alone wouldn't have been enough for this process. I had to be equally comfortable with Leadership Principles, communicating my thought process, adapting to follow-ups and explaining why I made specific technical decision.