Software Development Engineer II at Flipkart

How it went

What they would tell you

How to prepare

Round by round

1. 1

   Machine Coding90 min

   I was asked to design and implement a Buy Now Pay Later (BNPL) system using object-oriented design principles and relevant design patterns. In my approach, I used the Strategy Pattern, Singleton, and followed the Interface Segregation Principle. I wrote a clean LLD in Java, covering key components like credit evaluation, transaction handling, and user state management. I submitted the zipped code through a Google Form, and later that day, had a 30-minute discussion to walk through my design decisions and implementation.

2. 2

   DSA86 min

   This round typically includes two DSA questions, but since I completed both ahead of time, the interviewer gave me a third question as a bar raiser. I wrote my solution on a Google Doc, treating it like a whiteboard, and walked through test cases to explain my logic.

   • Binary Search based problem (don’t recall specifics)
   • Koko Eating Bananas
   • Search in Rotated Sorted Array (Bar Raiser Question)
3. 3

   System Design60 min

   Design the BookMyShow system focusing on search, seat availability, and third-party APIs (e.g., PVR). I began by listing out the functional and non-functional requirements to set the scope. Then I worked on designing the database schema, API endpoints, and a rough system flow. My main focus was on covering the core features like ticket search and booking. Where I Struggled:

   • Couldn’t properly design interaction with external vendors
   • Missed edge cases in seat availability updates, locking mechanisms, and caching
   • Went straight to ElasticSearch instead of a SQL-first approach
   • Was unsure how to reduce frequent hits to PVR’s API, or sync bookings across multiple platforms
   • Schema for events/movies wasn’t scalable, opted for parallel design without clear justification
4. 4

   System Design Re-evaluation75 min

   Same BookMyShow design with a focus on refining previous mistakes.

   • What’s the first API call when user lands on the page? What does the response look like?
   • How would you design schema for events vs. movies vs. shows?
   • Can existing schema support all use cases, or do we need parallel schemas?
   • How to minimize API calls to external ticket partners like PVR?
   • How to sync real-time bookings from other platforms into your system?
   • What to do with a SeatAvailability table that grows huge for every show? Approach: Improved over previous round, but still failed to answer key parts like real-time sync and optimal schema Design was somewhat improved, but still lacked scalability in schema and clarity on third-party integrations

What came up