Real questions asked at India's top consumer tech companies. Covers SDE, frontend, data, and product roles with round-by-round tips.
Practice with AI interviewer for any JD - starting at Rs 99
Swiggy, Zomato, and Flipkart each have distinct interview patterns. Know which to expect.
Compiled from 2024-2026 candidate reports on Glassdoor, Reddit, and community forums.
Tip: Start with a simple geospatial index (geohash or quadtree). Talk about rider state (online/busy/offline), load balancing across riders, and what happens when a rider declines. For senior roles, extend to multi-order batching. Draw a diagram, state your assumptions, and iterate.
Tip: Focus on working code first, polish later. Use a single state object or useReducer for filters. Debounce the search input. Handle loading and empty states. Interviewers watch how you structure components, not whether you finish every feature.
Tip: A heap per category is the textbook answer. For a streaming variant, mention approximate algorithms (count-min sketch). Discuss memory trade-offs and what happens at Flipkart scale (millions of products). Always ask clarifying questions before coding.
Tip: This is a Travelling Salesman variant. For small K (<=10), DP with bitmask works. For real-world scale, mention greedy heuristics + 2-opt. Interviewer cares more about how you reason about constraints than whether you code TSP perfectly.
setTimeout(() => console.log('a'), 0); Promise.resolve().then(() => console.log('b')); console.log('c');Tip: Output is c b a. Explain the call stack, microtask queue (Promises), and macrotask queue (setTimeout). Microtasks run before the next macrotask. Zomato loves this question because it reveals how deep your JS knowledge actually goes.
Tip: Key concepts: queueing (users enter a virtual waiting room), distributed locks on inventory, eventual consistency for cart updates, graceful degradation. Mention CDN for static pages, Redis for counters, and idempotency keys. Show you understand the difference between "available to view" and "available to buy."
Tip: Use STAR (Situation, Task, Action, Result). Pick a real example where you were accountable. Swiggy values engineers who own their bugs — don't blame teammates or vague "the API was slow" excuses. End with what you changed in your process.
Tip: Explain the tree diffing algorithm, why React uses keys (to track element identity across re-renders), and when Fiber was introduced (React 16) to enable incremental rendering. Bonus: mention useMemo/useCallback and when they actually help vs. hurt performance.
Tip: This is the classic "coin change" DP problem. State: dp[amount] = min coupons needed. Transition: dp[i] = min(dp[i], dp[i - coupon] + 1). Handle the "impossible" case. Flipkart often frames DSA as shopping problems — the structure is still classic DP.
Tip: Token bucket or leaky bucket. Discuss trade-offs (token bucket allows bursts, leaky smooths). Mention where rate limiting lives (gateway vs. service), how to handle distributed state (Redis), and what headers to return (X-RateLimit-Remaining). Bonus: adaptive limits based on user tier.
Tip: Break it down. Ask: which cities, which time slots, which restaurant tiers? Check recent app releases or A/B tests. Look at funnel metrics — is it fewer searches, fewer views, or fewer conversions? Don't propose solutions until you have hypotheses backed by data. PMs are graded on structured thinking, not jumping to fixes.
Tip: Show you've done homework. Flipkart's strength is tier-2/tier-3 India penetration, vernacular experience, and Big Billion Days scale. Mention a specific Flipkart product or initiative (PhonePe, Myntra, Shopsy) and connect it to why you want to join. Generic "I like your company" fails.
Tip: Use ROW_NUMBER() OVER (PARTITION BY city ORDER BY order_count DESC), then filter where rank <= 3. Interviewer may follow up with "what if two restaurants tie?" — then use DENSE_RANK(). Swiggy data roles test SQL thoroughly, especially window functions.
Tip: CSS Grid with repeat(auto-fit, minmax(240px, 1fr)) or clamp() for fluid typography. Container queries (@container) are now widely supported in 2026 and Zomato loves candidates who know them. Demo fluid spacing, aspect-ratio, and accessible focus states.
Tip: Quantify impact (users affected, latency reduction, revenue moved). Own the trade-offs you made. Be honest about what you'd do differently. Flipkart hiring managers use this to assess ownership and seniority. Practice this 3-4 times before the real interview — it's the most common "make or break" question.
Common questions from candidates preparing for Indian unicorn interviews.
Swiggy SDE interviews typically have 4-5 rounds: an online coding assessment (HackerEarth or similar), a DSA round focused on arrays, strings, trees, and graphs, a low-level design round (design a food delivery system, cart, or notification service), a system design round for SDE-2 and above covering scalability, caching, and distributed queues, and a hiring manager round focused on culture fit and past project impact.
Flipkart SDE interviews are considered medium-to-hard. Expect 2-3 DSA rounds with LeetCode medium/hard problems, one machine coding round (build a feature in 90 minutes, run and test), and one design round. Flipkart emphasizes production-quality code with edge cases, so practice writing clean, tested code, not just brute-force solutions.
Zomato frontend interviews cover JavaScript fundamentals (closures, event loop, promises, this binding), React deep-dive (hooks, reconciliation, performance), CSS (flexbox, grid, responsive design), one DSA round (mostly easy-medium), and one machine coding round where you build a small feature like an autocomplete, infinite scroll list, or a restaurant card component. Be ready to explain trade-offs.
Fresher SDE packages in 2026: Swiggy offers around Rs 20-28 LPA for top-tier campuses and Rs 12-18 LPA for off-campus. Zomato offers Rs 15-25 LPA range. Flipkart offers Rs 28-42 LPA for on-campus and Rs 18-28 LPA for off-campus. Product and design roles typically pay Rs 5-10 LPA more than SDE-1 at the same company.
Start with 150-200 LeetCode problems focusing on arrays, strings, trees, graphs, and dynamic programming. Learn low-level design patterns (Strategy, Observer, Factory) through building small systems. For senior roles, study system design fundamentals (caching, sharding, eventual consistency). Practice mock interviews to build speaking fluency and handle follow-up questions.
MockAce lets you paste a real Swiggy, Zomato, or Flipkart job description and our AI generates interview questions matched to that specific role. The AI plays the interviewer, evaluates your answers, asks follow-up questions, and returns a detailed scorecard. Practice as many times as you want at Rs 99 per session — cheaper than one hour of a human coach.
Browse live openings at top Indian unicorns on MockAce Jobs, pick one, and our AI generates interview questions from that exact job description. Instant scorecard and feedback.