Amazon SDE1 AUTA Interview Experience — 1.5 YOE | 2024 Batch | July 26

598
Sep 24, 2026

Sharing my Amazon SDE1 AUTA interview experience for anyone preparing for similar interviews.

Background: 1.5 years of experience, 2024 batch.

Timeline

  • 3 July: Received the survey form.
  • 17 July: Received the mail for onsite interview rounds, scheduled for 23 July.
  • I was told they would try to complete all rounds on the same day, depending on panel availability. Virtual interviews were also possible depending on availability.
  • There was no reimbursement for the onsite interview.
  • Reporting time was 9 AM. Around 15+ candidates were present, but the interview panels were late and interviews started around 12 PM.

Round 1 — DSA + Leadership Principles + Projects

Question 1

Given an array representing weights that need to be transported using a truck with maximum capacity k.

Only contiguous weights can be transported together.

Find the minimum time/number of trips required to transport all the weights.

I initially came up with a binary search on answer approach and explained the logic. The interviewer asked me to write the complete code.

Question 2

There are n cities and a list of possible roads. Each road has a cost cr.

Building a hospital in a city costs ch.

The goal is to make sure every city can access at least one hospital while minimizing the total cost.

I discussed the following:

  • If ch < cr, it is cheaper to build a hospital in every city.
  • Otherwise, for every connected component, we only need one hospital.
  • If a connected component has k cities, k - 1 roads are sufficient to connect all of them.
  • Therefore, if there are c connected components, the total number of roads required is n - c.

So the answer becomes:

c * ch + (n - c) * cr

Initially, I suggested using DSU to construct the connected components. The interviewer asked about the complexity, and I mentioned the extra logarithmic factor from DSU operations.

He then asked if there was a simpler approach.

I derived the above observation that a component with k nodes needs only k-1 roads, so we only need to determine the number of connected components.

The interviewer asked me to write the code.


Round 2 — DSA + CS Fundamentals + Projects

The DSA problem was approximately:

Given two arrays a and b and an integer k, select exactly k indices.

The score is:

sum(selected elements from a) * max(selected elements from b)

I initially started with a brute-force approach of around O(n^k).

Then I moved to a DP/knapsack-style approach with state based on n and k, giving approximately O(n*k).

Finally, I arrived at the greedy solution:

  • Pair the elements of a and b.
  • Sort according to b.
  • Iterate through the elements in decreasing order of b.
  • For the current index, its b value becomes the maximum b among the selected elements.
  • Maintain the best possible sum of k corresponding a values using a multiset.
  • Add the current a, remove the smallest element when necessary, and update the answer.

The final complexity was approximately:

O(n log n + n log k)

I was not asked to write the complete code, but I wrote it anyway since there was enough time.

After DSA, the interviewer went through my resume and asked questions around:

  • Kafka
  • ISR
  • Consistent hashing
  • Project-related discussions

I was able to answer most of these.

After the round, HR told me that I had been selected for the next rounds. However, the next interview panelists were in another meeting, so the remaining rounds would be conducted virtually.


Round 3 — DSA + Projects + Leadership Principles

This round had DSA, project discussion and Leadership Principles.

The DSA question was the standard Rotten Oranges problem.

I used the usual multi-source BFS approach.

The interviewer repeatedly asked whether the problem could be solved using "just BFS". I explained that the approach itself is BFS, but because multiple oranges are initially rotten, we need to initialize the BFS queue with all rotten oranges simultaneously — hence multi-source BFS.

There was also some grilling around Leadership Principles and project-related discussions.


Follow-ups

  • 4 August: I mailed HR asking for an update. I subsequently got my third round scheduled.
  • 11 August: Followed up again. HR replied that feedback was still pending.
  • 19 August: Followed up again.

I received the following response:

"Hiring plans for this position have changed and we will not be proceeding with interviews at this time."

And that's where the process ended.

Overall, it was a fairly long process from the initial survey to the final update.

Sharing this mainly for anyone preparing for Amazon SDE1 AUTA interviews, especially candidates from the 2024 batch / around 1–2 YOE.

Good luck to everyone preparing!

Comments (1)