Razorpay Interview Questions: 10 Real Q&A + Prep Tips (2026)
Curated Razorpay (Razorpay Software Pvt. Ltd.) interview questions with model answers.
Industry: Fintech. Difficulty: Hard. Hiring for: Software Engineer (SDE-1), Frontend Engineer, Backend Engineer, Data Engineer. Average package: ₹16-25 LPA (SDE-1), ₹30-45 LPA (SDE-2).
Selection process
- Resume screen (heavy on projects + GitHub)
- Online Coding Round — 2-3 problems on HackerRank, ~90 min
- Technical Round 1 — DSA + design (1 hour)
- Technical Round 2 — system design (for SDE-2+) or in-depth project (for SDE-1)
- Hiring Manager Round — culture + problem-solving
- Bar Raiser / Founders Round (case-by-case)
Real interview questions and answers
Technical: Implement an LRU cache.
Use a hashmap + doubly-linked list. Hashmap maps key → node; the list maintains usage order (most-recent at head, least-recent at tail). On get: move node to head. On put: if exists, update + move to head; if new and at capacity, evict tail, insert at head. O(1) get and put. Razorpay loves this question — practice the implementation cold.
Technical: Design a rate limiter.
Approaches: (1) Token bucket — tokens refill at a fixed rate, each request consumes one. (2) Sliding window log — store request timestamps, count those in the last N seconds. (3) Fixed window counter — simple but has burst issues at boundaries. For a payments system like Razorpay, token bucket per-API-key is standard. Discuss tradeoffs: memory vs accuracy vs implementation complexity.
Technical: How would you handle idempotency for a payment API?
Client passes an idempotency key (UUID) in the request header. Server checks Redis or DB: if key exists with completed response, return that response (don't re-process). If key exists with in-flight status, return 409 conflict. If new, lock the key, process, store result, release lock. TTL on the key (24h typical). This is core to Razorpay's actual architecture — they will dig deep on this.
Technical: Find the longest substring without repeating characters.
Sliding window: two pointers (left, right), a hashmap of char → last-seen index. Move right pointer; if current char is in the map and its index >= left, move left to (last index + 1). Update max length each iteration. O(n) time, O(min(n, alphabet size)) space. Razorpay tests fundamentals like this in round 1; expect to write the code and dry-run with an example.
Technical: Explain the difference between SQL transactions ACID properties.
Atomicity (all operations in a transaction succeed together or none do — rollback on failure). Consistency (DB moves from one valid state to another; constraints upheld). Isolation (concurrent transactions don't interfere — controlled by isolation levels: Read Uncommitted, Read Committed, Repeatable Read, Serializable). Durability (committed data survives crashes — via WAL/journal). Critical for any payments system.
Technical: Difference between optimistic and pessimistic locking.
Pessimistic: lock the row before reading (SELECT FOR UPDATE), block others. Use when conflicts are likely. Optimistic: read without lock, but check a version number on update — if version changed, reject and retry. Use when conflicts are rare. Razorpay's payment flows use pessimistic locking on order rows during capture; optimistic for high-throughput merchant settings.
Behavioral: Tell me about a time a production bug was your fault.
Pick a real one. Structure: what broke, how it was detected (monitoring? user report?), your immediate response (rollback, hotfix?), root cause, what you changed in the team's process to prevent repeats. Razorpay interviewers value ownership over avoiding blame. "I added a runbook step + a monitoring alert" beats "I was more careful next time."
Behavioral: How do you decide between multiple technical approaches?
Walk through your decision framework: (1) Read the constraints (latency, scale, team skill). (2) List 2-3 viable options. (3) Pick the simplest that meets constraints — bias toward boring tech. (4) Identify the one risk in your choice and how you'd test for it. Razorpay's engineering culture rewards judgment over cleverness; show you're calibrated.
HR: Why Razorpay over a bigger product company?
Specific reasons that signal you've done research: payments is a uniquely hard domain (regulatory, scale, latency-critical), Razorpay's product breadth (PaymentGateway, RazorpayX, Capital), the engineering blog (read it), or the autonomy that mid-stage startups offer over FAANG. Avoid "I want to learn fast" — too generic.
HR: Where do you see yourself in 3 years?
Honest version: deepen in one technical area (specify which — distributed systems, frontend infra, ML, etc.), grow toward tech lead responsibilities. Avoid "I want to be a manager" unless you actually do — Razorpay tracks IC and management as separate paths.
Preparation tips
- Razorpay leans heavy on system design even for SDE-1. Practice 2-3 design problems specifically: rate limiter, payment processor, URL shortener.
- Read Razorpay's engineering blog before your interview (engineering.razorpay.com). Reference an article in your "why Razorpay" — it lands.
- Idempotency, ACID, and transaction handling come up in almost every backend interview. Don't fake understanding.
- For frontend roles: be ready on React performance (virtual DOM, memoization, code-splitting), accessibility, and at least one CSS-in-JS approach.
- Bring questions that show you understand the business: "How does Razorpay think about international expansion vs domestic depth?" beats "What's the tech stack?"
Common questions about Razorpay interviews
What is the typical SDE-1 package at Razorpay?
SDE-1 fresher / 0-2 YoE typically gets ₹16-25 LPA fixed + ESOPs + joining bonus. Total CTC range ₹20-32 LPA depending on negotiation and prior offers.
Does Razorpay hire from tier-3 colleges?
Yes, but the bar is higher. Tier-3 candidates typically need standout signals — strong GitHub, hackathon wins, previous internship at a product company, or ICPC/Kaggle background. The interview itself is identical regardless of college.
How long is Razorpay's interview process?
Resume screen to offer typically 3-5 weeks. Process: OA (90 min) → 2 technical rounds (1 hour each) → HM round → bar raiser. Razorpay is known to be responsive — feedback usually comes within 5 days of each round.
What's the work culture like at Razorpay?
High autonomy, strong engineering culture, hybrid (3 days office in Bangalore HQ). Heavy on documentation and async communication. Engineering blog and open-source contributions are encouraged. Reviews are calibrated; promotions are based on impact, not tenure.