Indian Unicorn

Razorpay

Free first round · 20 min · no credit card

FAANG-comparable engineering bar. Payments-domain rigor + idempotency probing.

Interview philosophy

How Razorpay actually interviews

Razorpay sits at the top of the Indian unicorn engineering bar — interviewers consistently benchmark against FAANG, and candidates report a process that is closer to Stripe or Google than to a typical Indian product company. The loop is rigorous and slow: a short DSA round is followed by a 90–120 minute machine-coding round where you build a fully running, extensible system with proper design patterns, then a deep HLD round, a past-experience deep-dive, and a hiring-manager round that re-tests behavioural and design judgment.

The distinctive vibe is payments-domain seriousness: idempotency, exactly-once semantics, reconciliation, and financial correctness come up in nearly every senior round, and getting these wrong is fatal even if your code compiles and runs. Interviewers tend to be senior Razorpay engineers and tech leads who will dig into your past projects with the same depth as the design problem itself — vague ownership claims do not survive contact.

Culturally, Razorpay rewards first-principles thinking and quiet confidence. Loud framework-reciting and over-rehearsed STAR stories tend to under-perform; candidates who can re-derive the answer from fundamentals and admit what they don't know advance further. Process is structured but humane — recruiters communicate well and feedback (when given) is usually substantive.

A "hire" at Razorpay senior SDE means: writes production-quality code with explicit failure-mode handling, designs for idempotency and exactly-once semantics by default, can defend trade-offs from first principles, and clears a financial-correctness bar.

Cultural pillars

What they're measuring you on, beyond the right answer

The values interviewers probe for. Each pillar is what they ask about, plus how they ask it.

Customer Delight

Decisions and actions are guided by a commitment to deliver superior customer experience; customer obsession is the first of Razorpay's six principles.

How they probe · Asks for a specific story where you went beyond the spec because the customer outcome demanded it; pushes on what data made you change course.

Make a Difference

Razorpay hires people who want to leave the system better than they found it — small fixes count, but signing up for big-impact problems counts more.

How they probe · Asks: 'what was the biggest improvement you drove that wasn't on your JD?'; probes for owned outcomes vs. assigned tasks.

Engineering Excellence (the best build the best)

Razorpay self-positions as a company where the engineering bar is FAANG-comparable; financial correctness, idempotency, and scale are non-negotiable.

How they probe · In machine-coding and HLD, looks for clean abstractions, idempotency awareness, and explicit treatment of failure modes — not just a working happy path.

First-Principles Ownership

Razorpay leans heavily on first-principles thinking — candidates are expected to derive answers from fundamentals rather than recite frameworks.

How they probe · Will deliberately push back on a 'standard' answer to see if you can re-derive it; asks 'why' three or four levels deep.

Scale Bigger

Razorpay is the largest payments platform in India by volume and expects engineers to think in millions of TPS, not thousands.

How they probe · HLD round explicitly asks how the design changes at 10x and 100x; probes for backpressure, sharding, and partial-failure handling.

The full loop

Round-by-round, in the order they actually run

Reported pattern from candidate write-ups. Eliminating rounds are the ones where a single bad signal ends the loop.

01
Recruiter screen30 minNon-eliminating
career story, motivation for fintech, comp expectations, role fitment
02
DSA screen60 minEliminating
1–2 medium/medium-hard DSA problems on a shared editor; optimal solution + complexity
03
Machine coding120 minEliminating
build a fully-running, extensible mini-system (e.g. EMI calculator, payment router) with proper OOP, design patterns, and tests
04
System design (HLD)75 minEliminating
design a payments-domain system end-to-end; idempotency, retries, reconciliation, scale to 10x/100x
05
Past-experience deep dive60 minEliminating
drill into 1–2 of your past projects: architecture, trade-offs, failures, your specific contribution
06
Hiring manager60 minEliminating
behavioural + values fit + culture; revisit a design choice from earlier rounds

Real questions, by round type

What candidates were actually asked

Curated from interview reports and company write-ups. Practise against any of these in a live mock.

SDE · Senior

Coding

Implement an in-memory LRU cache with O(1) get/put.
Design a rate limiter (token bucket / sliding window) — coded, not just discussed.
Find the maximum path sum in a binary tree.
Given a stream of integers, return the running median.
Implement a thread-safe bounded blocking queue.
Parse and evaluate an arithmetic expression with parentheses and precedence.

Machine coding

Build an in-memory loan EMI calculator that supports multiple loan types and partial pre-payments.
Build a parking-lot system with multiple vehicle types, pricing strategies, and floor allocation.
Build a payment-router that picks the best gateway given cost, success-rate, and bank-down rules — make it extensible to new gateways.
Build a splitwise-style expense tracker with settlement minimisation.

System design

Design Razorpay Payment Pages end-to-end: how does an idempotent /payments/create API work under retries and gateway timeouts?
Design a payouts system that must guarantee exactly-once disbursement across bank failures.
Design a reconciliation pipeline that matches our internal ledger to bank settlement files daily.
Design a webhooks delivery system with at-least-once delivery, retries, and exponential backoff.
Design a fraud / risk-scoring service that has to respond in <50ms during checkout.

Behavioral

Tell me about the biggest production incident you've owned end-to-end — what did you change after?
When did you push back on a product decision and what was the outcome?
Describe a project where you had to make a trade-off between consistency and availability — what did you choose and why?
When have you raised the engineering bar of the team around you?

Product · Senior

Product sense

Should Razorpay build its own UPI consumer app? Walk me through the strategy.
Design a product to help senior citizens transact online safely.
How would you reduce payment failure rate on Razorpay checkout by 100 bps?
Design a washing machine for blind users — explain your structuring.
A merchant tells us reconciliation is broken. What do you do in week 1?

Strategy

Razorpay X has 40% of small-business neobanking. How do you sustain the lead against an incumbent bank that wakes up?
Tell me about a time you used data to change a leadership decision.
What is the North Star metric for Razorpay Capital and why?

Behavioral

Tell me about a product launch that failed and what you learned.
When did you say no to engineering to protect the customer?

What rejects you · what advances you

The two patterns that decide every loop

Red flags

Vague ownership in past-project stories: 'we did X' instead of 'I did X' — Razorpay deep-dives expose this fast.
Memorised system-design templates without understanding idempotency, retries, or reconciliation in a payments context.
Job-hopping every 12 months without a clear narrative; Razorpay prefers 2y+ tenure for senior roles.
Treating financial-correctness questions as edge-cases ('we'll handle that later') — for Razorpay these are the design.
Bad-mouthing previous employers or revealing confidential information from prior roles.
Negotiating purely on hike% without engaging with role/scope fit.

Advance signals

Re-derives standard answers (CAP, idempotency keys, sharding strategies) from first principles rather than reciting them.
Proactively surfaces failure modes — gateway timeout, network partition, double-spend — before being asked.
Writes machine-coding solutions that are testable, extensible, and use clean OOP without over-engineering.
In past-project deep-dives, can articulate the trade-off they would make differently today.
Asks sharp clarifying questions on consistency, latency budget, and scale before designing.
Demonstrates fintech/payments domain awareness — RBI norms, UPI architecture, settlement cycles.

Don't do

Don't hand-wave over idempotency, retries, or duplicate-prevention in any payments design.
Don't bring competitor confidential data (Stripe/Adyen internals, etc.) into your answers.
Don't use over-engineered design patterns in machine coding — interviewers explicitly penalise unnecessary abstraction.
Don't skip writing tests in the machine-coding round even if not explicitly asked.
Don't argue with the interviewer when pushed back on; engage with the critique.

Compensation

Base salary bands by level

Junior

₹18–28L

Annual base

Mid

₹32–55L

Annual base

Senior

₹55–90L

Annual base

Staff+

₹90–150L

Annual base

Heavy ESOP component for senior+ (often 20–35% of TC over 4y vest). Levels.fyi median total comp: SE ~₹22L, SSE ~₹41L, Lead ~₹54L (India). Joining bonus typical at senior+. No retention bonus culture.

Process

What to expect after each round

Typical timeline

~3 weeks

Recruiter-screen → offer

Reapply window

6 months

After a final-round rejection

Feedback practice

Substantive feedback is occasionally given by the recruiter on rejection at later rounds; rare at the screen stage.

“Thanks for your time today — your recruiter will reach out within 3–5 business days with next steps or feedback.”

Verified profile

Last verified Mon Apr 27 2026 00:00:00 GMT+0000 (Coordinated Universal Time) · 10 sources

View sources