PayU Campus Interview: Six-Stage Candidate Account
A PayU campus candidate reported an online test, four technical interviews, and an HR interview covering DSA, CS fundamentals, and puzzles.
on this page§ 08
The six-stage process in the reviewed account
The GeeksforGeeks account describes a PayU campus process for a Software Engineer position. It records an online test, four technical interviews, and an HR interview. The source gives detailed questions and durations, but it does not establish a company-wide hiring design, scoring philosophy, or current cycle.
The reported online assessment mixed technical and coding questions. Four technical interviews followed, each with a different mix of projects, data structures, recursion, networking, databases, operating systems, and design. The final reported stage was an HR interview lasting about twenty minutes. Use the account as a question bank and confirm current structure from your campus notice or PayU's live application instructions.
For preparation, map each sourced question to the underlying skill. That keeps the account useful even if the current round count, platform, or exact prompts have changed.
Breakdown of technical questions and rounds
| Round | Duration | Focus Areas | Specific Questions |
|---|---|---|---|
| Online Test | N/A | Coding & Technical | 15 technical questions, 5 coding questions |
| Technical Round 1 | 30 min | Projects & DSA | LCA in BST, remove characters of string2 from string1 in O(N), modified rod-cutting DP |
| Technical Round 2 | 45 min | Data Structures & Puzzles | AVL tree implementation, array pair-finding, 16 horses puzzle, web sessions/cookies |
| Technical Round 3 | 30 min | Recursion | English-dictionary coding challenge with progressively complex test cases |
| Technical Round 4 | 40 min | CS Fundamentals & Design | KMP preprocessing, HTTP vs HTTPS, OS paging, singleton design pattern, mutual friends feature design |
| HR Round | 20 min | Behavioral | Self-introduction, motivation, challenging project, strengths/weaknesses |
The candidate reported solving all five coding questions and about ten of the fifteen technical questions before being shortlisted. That is the candidate's result, not a published cutoff. Technical Interview 1 covered projects, a modified lowest-common-ancestor task in a BST, a modified rod-cutting problem, and string filtering with a required linear-time, constant-space solution. The useful preparation lesson is to derive variants rather than memorise one standard implementation.
Technical Interview 2 asked the candidate to implement AVL-tree insertion and deletion using maps, discuss web sessions and cookies, solve array-pair and subset variants, and work through a horse-ranking puzzle under stated race constraints. Technical Interview 3 used an English-dictionary problem with increasingly complex inputs to probe recursion, base cases, and adaptation. These details are directly reported; the source does not tie them to PayU's industry or guarantee repetition.
Technical Round 4 covered KMP preprocessing, HTTP versus HTTPS, network security, database normalisation, operating-system paging, the Singleton pattern, and feature-design prompts involving mutual or top friends. This list comes from the candidate account. The source does not publish the scoring or identify which answers determined progression.
How to reason through the reported technical work
The useful part of a historical question list is the reasoning it exposes. For the modified BST task, first clarify whether both target keys are guaranteed to exist, whether the tree has unique keys, and what the requested modification changes from a standard lowest-common-ancestor problem. State the invariant that allows the search to move left or right. If the modification invalidates that invariant, say so before selecting another traversal.
For the string-filtering task, clarify whether the input buffer may be changed. A linear-time, constant-auxiliary-space solution can scan the second string to build membership only when the character domain permits a fixed-size representation, or it can repeatedly test membership at a different cost. The candidate should connect the claimed complexity to the actual character model instead of repeating O(N) without defining what N covers.
For AVL insertion and deletion, explain the height invariant, how balance is measured after a change, and why the rotation case follows from the affected path. The source says maps were involved, but does not give the full interface. Ask whether the map stores keys, nodes, parents, or auxiliary metadata rather than silently inventing the contract.
For KMP preprocessing, define what each prefix value represents and trace how fallback reuses a previously matched border. A correct answer should distinguish preprocessing from the later search and show how the fallback prevents restarting every comparison at the beginning. Use a small pattern chosen during practice, not an unverified claim about the exact interview input.
The mutual-friends design prompt should begin with requirements. Clarify whether friendship is symmetric, whether blocked or private relationships exist, what “top” means, and what consistency is required. Then choose a representation and query plan. Do not add distributed-system scale unless the interviewer supplies it.
For modified rod cutting, define what a cut may produce, whether every length has a value, and whether the full rod must be consumed. Write the recurrence from those rules and identify repeated subproblems before selecting memoisation or tabulation. A changed constraint may invalidate the standard recurrence, so derive it rather than naming dynamic programming as a slogan.
For the recursive dictionary task, the source says the interviewer increased input complexity but does not publish enough of the prompt to reconstruct it safely. The preparation lesson is to state the base case, show that each recursive call makes measurable progress, and discuss repeated work, stack depth, and invalid input. Do not invent sample words or test cases and attribute them to PayU.
Preparation mapped to the sourced questions
Use the account to build skill buckets, not to predict a current sequence.
Master standard DSA with variants
This account contains modifications of familiar problems. Practise stating the original invariant, identifying what the changed constraint breaks, and rebuilding the solution from that point. For every complexity claim, define the input size and auxiliary storage being counted.
Brush up on CS fundamentals
Technical Interview 4 covered HTTP versus HTTPS, network security, database normalisation, OS paging, and the Singleton pattern. Review these subjects by explaining mechanisms and trade-offs, then writing small examples where code is relevant. The source does not rank which topic affected the outcome.
Practise constrained-reasoning puzzles
The source proves that this candidate faced a horse-ranking puzzle, not that PayU has a stable puzzle policy. For a constrained puzzle, list the permitted observations and operations, derive what each step establishes, and avoid assuming a timer, comparison, or tie rule that the prompt has not supplied.
Prepare projects and HR scenarios
The account includes project discussion in technical and HR stages. Be ready to explain the problem, your contribution, technical decisions, constraints, and a real challenge. Build your motivation answer from the current role description and PayU's own pages, without inventing scale or product claims.
Red Flags: What NOT to Do in Your PayU Interview
- Ignoring space complexity: The first technical interview explicitly required linear time and constant auxiliary space for the string-filtering task. State how the input representation affects whether that requirement is achievable.
- Rote learning algorithms: A modified LCA or rod-cutting task can invalidate a memorised template. State the invariant and adapt it to the supplied constraint.
- Skipping CS fundamentals: Round 4 in this account included OS paging, database normalisation, networking, and design patterns. Review the named mechanisms and trade-offs without inventing how the panel weighted them.
- Inventing puzzle rules: Restate the permitted comparisons, race conditions, and target before solving. A familiar puzzle may have different constraints.
- Vague HR answers: Use real project challenges and current role details. Do not rely on generic company praise or unsupported claims about PayU's business.
Real-world data points
- Six stages: Online test, four technical interviews, and one HR interview in the candidate-reported process.
- 15 technical + 5 coding questions: Format of the online test (candidate-reported).
- 30 min: Duration of Technical Round 1 and Round 3 (candidate-reported).
- 45 min: Duration of Technical Round 2 (candidate-reported).
- 40 min: Duration of Technical Round 4 (candidate-reported).
- 20 min: Duration of the HR round (candidate-reported).
- O(N) time and constant space: Required complexity for the string manipulation problem in Round 1 (candidate-reported).
FAQ
How many rounds are there in the PayU on-campus interview process in 2026?
The reviewed account reports an online test, four technical interviews, and an HR interview. That is six evaluation stages in this campus account, not a verified current PayU-wide pattern.
What kind of coding questions are asked in PayU technical interviews?
The reviewed account reports BST and AVL-tree questions, modified rod cutting, KMP preprocessing, recursion, array tasks, web sessions, networking, databases, operating systems, design patterns, and a horse-ranking puzzle.
What topics are covered in the PayU online test?
In this account, the online test contained fifteen technical questions and five coding questions. The candidate reported solving all coding questions and about ten technical questions, but the source does not publish a shortlist cutoff.
Sources & credits
Sources and review notesreviewed 11 Aug 2026
Official notices, candidate reports, offer documents, and editorial practice questions carry different confidence levels. The visible source list lets you inspect the evidence instead of relying on a blanket verification badge.
topic cluster
More resources in Interview Questions
Use the category hub to browse similar questions, exam patterns, salary guides, and preparation resources related to this topic.
paid contributor programme
Sat this this year? Share your story, earn ₹500.
First-person experience reports help future candidates prep smarter. We pay verified contributors ₹500 via UPI per accepted story with byline.
Submit your story →ready to practice?
Take a free timed mock test
Put what you learned into practice. Our mock tests match the 2026 pattern with timer, navigator, reveal, and score breakdown. No signup.