My Juspay Interview Experience — SDE 2026

(image taken from web)

I applied for the SDE-1 Backend role at Juspay through their careers page without much expectation. To my surprise, the HR team called me within days and scheduled the entire process for their Kolkata office. What followed was one of the most intense and mentally exhausting interview loops I’ve ever experienced.

Online Assessment (OA) — 1 Hour

• 16 MCQs covering CS fundamentals, Operating Systems, Computer Networks, Physics, and Maths.
• Nothing too tricky if you’ve revised your basics, but the mix of subjects kept you on your toes.

Online Coding Round — 1.5 Hours

• 3 questions were there (keep in mind, you cannot go back a question) one was fairly easy (sliding window), one medium and one hard (fenwick Tree).
• If you are able to get 2 questions right, you will make through this round.

Hackathon Part A — 3 Hours

• The infamous “Tree of Space” problem (the exact same one that’s been floating around for years).
• You can read the original problem here: **Tree Of Space**
• I had practiced it before, so I managed to implement the node structure, locking/unlocking with proper parent-child checks, and handled all edge cases within time.

Elimination Round (New Round) — 1 Hour

• Juspay recently added this elimination round. The interviewer gave me only one LeetCode problem: **Open the lock**
• I had to: Explain brute force, Optimise with BFS, Do time complexity analysis and write the pseudo-code.
• No follow-ups. Just this one question. Cleared this round comfortably too.

Hackathon Part B — Full Day (11 AM — 7 PM)

• Assigned a mentor on Slack + a meeting link (used only when necessary).
• Task: Add multi-threading to the “lock” function of the earlier implemented “Tree of Space” problem with granular locking on the path. Of-course, you must design a solution keeping in mind that it’s fit for unlock and upgrade functions too in future.
• Rules: No libraries allowed. Implement everything from scratch.
• Questions fired non-stop: OS concepts (locks, deadlocks, race conditions, contention), dry runs, edge cases, real-time simulation.
• The interviewers were really very friendly throughout. He guided me through tricky parts without spoon-feeding, and we had good back-and-forth discussions on different locking strategies (fine-grained per node/path and coarser approaches).
• At the very end, we shifted to a relaxed technical discussion on my resume and projects. He asked questions about my past work and projects.

Systems Interview (Second-Last Round) — 3 Hours

• This was supposed to be a Systems round taken by a senior manager + another senior engineer. Before every round the HR will give you a detailed list of topics to prepare and for this round the list was, OS, CN, COA and System Design Concepts. I prepared exactly that.
• “Murphy’s Law”…. they dropped a Low-Level Design (LLD) bomb (i was least prepared for this). Problem was variation of traditional “Parking Lot” problem :-

“There’s a multi-storey shopping mall parking problem. 5 floors, each floor has sections A, B, C, D. Each section has 10 car slots. Each car slot can also park 4 motorcycles. Fare is ₹20 per hour. The system should suggest the best available spot for any vehicle. Implement park(), unpark(), calculate fare, etc. Make it a complete real-world simulation.”

• They dont’ just see your solution instead they focus more on how you reached there and how you thought about the edge cases while writing code. We have to implement the most efficient way of assigning and clubbing bikes to the spots such that most number of cars can be parked.
• My code was working fine but then the interviewers asked me that there’s one edge case missing, i took around 5–10 minutes, gave some dry runs and found that one test case. After few thoughts i was able to solve it but indeed it was not the most optimised solution. The case was a mix of few un-parking and parking operations.
• The round ended at the 3-hour mark. I knew I had underperformed.
• After one week, I got the rejection email. No cultural round for me 😿

Juspay’s process is genuinely different and tests real engineering and problem solving depth. Even though I got rejected, I respect the process a lot.

A lot of these interviews are dependent upon interviewing style, except the Tree of Space problem, not a lot of repeatability. This is my personal experience from early 2026. Interview processes can evolve, so take this as one data point. Shared purely for learning and to help fellow candidates prepare.

Best of Luck to any future candidates!

Let’s Connect

You are most welcome to follow me here on Medium. In Addition, feel free to check out:

• My portfolio
• My LinkedIn Profile