Coursera Backend Engineer interview questions

3 Backend Engineers have shared notes from their Coursera interviews, with the questions they were asked in each round. That is 23% of the 13 Coursera interview notes on Screna AI.

3 Backend Engineer interview notes

Coursera·Backend Engineer·Technical Phone Screen

Jun 2026
Coursera backend engineer interview with a voting algorithm problem. Two parts to the same question, one straightforward and one that actually requires thinking through the elimination logic carefully.
  • Given ballots represented as a dictionary mapping ranked candidate tuples to voter counts, implement a popular vote function that returns the candidate with the most first-place votes.
  • Using the same ballot structure, implement a ranked-choice voting algorithm that eliminates the last-place candidate each round and redistributes their votes until someone clears 50%.

“Pretty mechanical once you see it.”

View Post

Coursera·Backend Engineer·Technical Phone Screen

May 2026
Coursera backend interview with a pretty involved algorithmic problem around Instant-Runoff Voting. The question had enough edge cases that you really had to think through the procedure carefully, and clarifying tie-break behavior with the interviewer was part of the exercise.
  • You're given a map of election ballots where each key is a tuple representing a voter's full candidate ranking (most preferred first) and each value is the count of voters with that exact ranking. Implement the Instant-Runoff Voting algorithm: repeatedly tally first-choice votes, declare a winner if someone exceeds 50% of remaining votes, otherwise eliminate the last-place candidate and redistribute their ballots to the next ranked choice. What are the tie-break rules for elimination and for the majority check?

“The redistribution logic is where I got a bit tangled.”

View Post

Coursera·Backend Engineer·Technical Phone Screen

Apr 2026
Coursera backend engineer interview with a coding question around election ballot processing. Pretty clean problem on the surface but the tie-breaking condition is the kind of thing you can easily forget to handle under pressure.
  • You're given a map where each key is a tuple representing a voter's full candidate ranking (most preferred first) and each value is how many voters submitted that exact ranking. Find the candidate with the most first-choice votes. If there's a tie, return the lexicographically smallest name among the tied candidates.

“The core aggregation part is straightforward, just pull index 0 from each tuple and sum up the counts.”

View Post