Flipkart Interview Questions 2026: SDE Rounds, System Design, and Culture Fit

Flipkart SDE interview process (SDE-1/SDE-2/SDE-3):

1. Online Assessment (for campus and some off-campus drives): 2 coding problems (90 minutes, LeetCode medium-hard difficulty), sometimes a technical MCQ section.

1. Machine Coding Round (unique to Flipkart and a few other Indian unicorns): 90 minutes to 2 hours; build a working mini-application or feature in your language of choice. Judged on: code quality (SOLID principles, clean architecture, no god classes), completeness (core functionality works), edge case handling, and testability.

1. Technical Coding Rounds (2 rounds, 45-60 minutes each): LeetCode medium-hard problems; graphs (course schedule, number of islands), trees (LCA, serialise/deserialise), DP (knapsack, LCS), and arrays (merge intervals, trapping rain water).

1. System Design Round (SDE-2 and above): 'Design Flipkart's product search system', 'Design Flipkart's order management', 'Design a rate limiter'. 45-60 minutes.

1. HR / Culture Fit Round: Flipkart values: 'audacity', 'bias for action', 'customer centricity'. Align your STAR stories to these values.

The Machine Coding Round is Flipkart's most distinctive element:

What is tested: your ability to write production-quality object-oriented code under time pressure. Not algorithmic cleverness; clean design that another engineer could maintain.

Practice problems (in order of difficulty):

1. Parking Lot: vehicles of different sizes, multiple floors and spots, spot allocation and release
2. Library Management System: books, members, borrow/return, fine calculation, search
3. Snake and Ladder game: dice, snakes, ladders, multiple players, game state management
4. Movie Ticket Booking: screens, shows, seats, concurrent booking, payment
5. Ride Sharing System: drivers, riders, location matching, trip management, pricing
6. Inventory Management: products, warehouses, quantity tracking, reorder alerts

Approach during the round:

1. Spend 10-15 minutes designing on paper or in comments: entities, their attributes, relationships, and key methods
2. Agree the design with the interviewer before writing any code (this avoids wasted implementation)
3. Implement one feature at a time; make it work before adding the next feature
4. Write unit tests for core logic if time permits
5. Use meaningful names at every level: RideMatchingEngine not RME; availableDrivers not AD

Code quality checklist:

• Single Responsibility Principle: each class has one reason to change
• No god classes with 10+ methods and 15+ fields
• No nested ternaries or deeply nested conditionals
• Meaningful method names that describe what the method does

Flipkart India most common coding topics:

Arrays:

• 'Merge K sorted arrays' (min-heap: add first element from each array; pop min, add next from same array)
• 'Find the Kth largest element' (quickselect or min-heap of size K)
• 'Rotate an array by K positions' (reversal algorithm: reverse whole, reverse first K, reverse rest)
• 'Find maximum sum of a subarray of size K' (sliding window)

Trees:

• 'Vertical order traversal of a binary tree' (BFS with column tracking)
• 'Serialize and deserialize a binary tree'
• 'Lowest common ancestor'
• 'Binary tree zigzag level order traversal'

Graphs:

• 'Number of islands' (BFS/DFS)
• 'Course schedule II' (topological sort)
• 'Minimum spanning tree' (Kruskal's or Prim's: less common but appears for SDE-2)

Dynamic Programming:

• 'Edit distance'
• '0/1 Knapsack'
• 'Maximum product subarray' (track both max and min at each position)
• 'Longest increasing subsequence' (O(n log n) with binary search)

Flipkart system design questions:

1. 'Design Flipkart's product search system': Search API: query parsing, spell correction, synonym expansion. Indexing pipeline: product attributes indexed into Elasticsearch cluster; updated via Kafka events when products change. Ranking: ML model (user personalisation, popularity, inventory level, seller reputation, price competitiveness). Filtering: faceted search (category, brand, price range, rating) implemented as Elasticsearch aggregations. Caching: hot queries cached in Redis with 5-minute TTL (80% of search traffic is for the top 5% of queries).

2. 'Design Flipkart's order management system': State machine: placed, paymentpending, paymentconfirmed, processing, shipped, outfordelivery, delivered, cancelled, returned. Each state transition publishes an event to Kafka. Downstream services (inventory, seller, logistics, notification) consume the event and update independently. Idempotency: each state transition checks the expected previous state before applying.

3. 'Design a product recommendation system for Flipkart': Real-time signals: what the user is currently browsing (session context). Collaborative filtering: users with similar purchase history also bought X. Content-based filtering: products similar to what the user has viewed. Popularity-based fallback: top-selling in the user's category when no personalisation data exists. Serving: pre-computed embeddings cached in a feature store, ranking model served at low latency.

Flipkart total compensation (2024-25, Bangalore):

SDE-1: 25-35 LPA base + 10-20 LPA RSUs annualised = 35-55 LPA TC SDE-2: 35-55 LPA base + 20-35 LPA RSUs annualised = 55-90 LPA TC SDE-3 (Senior SDE): 55-75 LPA base + 35-60 LPA RSUs annualised = 90-130 LPA TC Staff SDE: 75-100 LPA base + 60-90 LPA RSUs = 130-200 LPA TC

Note: Flipkart became a Walmart subsidiary in 2018; RSUs are Flipkart pre-IPO shares, not Walmart shares. The liquidity event for these shares depends on a Flipkart IPO (which has been anticipated for several years but had not occurred as of mid-2026). Pre-IPO shares carry more liquidity risk than publicly traded RSUs.

Culture: Flipkart is known for a fast-paced, execution-focused culture. 'Audacity' (aiming at big, bold problems) and 'bias for action' (acting on imperfect information rather than waiting) are core values. Engineers at Flipkart frequently cite the scale and complexity of the technical challenges as the main draw. Bangalore headquarters; most SDE roles require on-site presence with limited WFH flexibility.