Fivetran SDE-2 interview review

fivetran logo
fivetran
SDE-2
June 12, 20252 reads

Summary

I interviewed for an SDE-2 role at Fivetran. After an online assessment, I proceeded to a technical round where I solved the Longest Increasing Subsequence problem and a problem conceptually similar to LeetCode 134: Gas Station. Despite solving both problems, I was rejected after the first technical round.

Full Experience

Position: SDE-2 Applied via: Direct application on Fivetran careers page for SDE role Language: Java only Status: Rejected after Round 1

Online Assessment (HackerRank)

The online assessment consisted of a single coding question on HackerRank, with Java as the only allowed language. The problem was based on managing employee scores. I approached it using a Heap-based solution, and only 2–3 test cases failed. Despite the partial failure, I was contacted by a recruiter shortly afterward. He informed me that the interview process would consist of two rounds, and I would proceed to the second round only if I cleared the first.

Round 1 (Technical Interview – 1 Hour)

Part 1: Longest Increasing Subsequence

The first round of the technical interview lasted an hour and was divided into two parts, both focused on data structures and algorithms. In the first part, I was asked to solve the Longest Increasing Subsequence (LIS) problem. I began by explaining the approach conceptually, starting with a decision tree and highlighting overlapping subproblems, which naturally led to a dynamic programming solution with a time complexity of O(n²). We did a dry run together on a sample input to ensure the logic was sound, and then I coded the solution. It passed all test cases. I was expecting a follow-up question to optimize the solution to O(n log n) using binary search and patience sorting, but that wasn’t asked.

Part 2: Aladdin and the Magic Carpet

In the second part, I was given a problem called Aladdin and the Magic Carpet, which was conceptually similar to Leetcode 134: Gas Station. At first, I had trouble understanding the problem, but the interviewer clarified it for me. I then implemented a greedy approach. My initial solution failed some test cases due to a bug in how I calculated the minimum fuel required. After identifying and fixing the issue, the updated solution passed all test cases successfully.

Outcome

I felt the interview round went well overall — both problems were solved with correct logic and fully working code after debugging. The interviewer was cooperative, and I was confident about my performance by the end of the session. Naturally, I was expecting to hear back for the next round, but after a few days of silence, I decided to follow up with the recruiter.

To my disappointment, the recruiter replied that the feedback was not positive :) and I wouldn’t be moving forward in the process. There was no detailed explanation provided.

Interview Questions (2)

Q1
Longest Increasing Subsequence (LIS)
Data Structures & AlgorithmsMedium

I was asked to solve the Longest Increasing Subsequence (LIS) problem. I began by explaining the approach conceptually, starting with a decision tree and highlighting overlapping subproblems, which naturally led to a dynamic programming solution with a time complexity of O(n²). We did a dry run together on a sample input to ensure the logic was sound, and then I coded the solution. It passed all test cases. I was expecting a follow-up question to optimize the solution to O(n log n) using binary search and patience sorting, but that wasn’t asked.

Q2
Aladdin and the Magic Carpet (Gas Station Variant)
Data Structures & AlgorithmsMedium

I was given a problem called Aladdin and the Magic Carpet, which was conceptually similar to Leetcode 134: Gas Station. At first, I had trouble understanding the problem, but the interviewer clarified it for me.

Discussion (0)

Share your thoughts and ask questions

Join the Discussion

Sign in with Google to share your thoughts and ask questions

No comments yet

Be the first to share your thoughts and start the discussion!