Flipkart Data Scientist interviews, per candidate reports, combine ML breadth, a case study, a hands-on modeling task such as reranking, and a hiring manager round on past projects. SQL and statistics also appear.
$58K
Avg. Base Comp
$72K
Avg. Total Comp
4 rounds
Typical Rounds
Not reported
Process Length
Three of the four candidates describe a four-round loop. It runs from an ML breadth round to a case study, then a hands-on modeling task, then a hiring manager conversation. The fourth candidate describes a multi-round process built around SQL and statistics, with no round count given.
The ML round is where depth matters most. One candidate says the round centered on derivations rather than just formulas, with notation expected. They name logistic regression, bias-variance, overfitting and underfitting. Another lists KNN, linear regression, precision, recall and F1, regularization, ensembles, transformers and dropout. A third says the round leaned toward deep learning.
The case and design round is framed around e-commerce problems. Candidates mention pricing optimization for mobile phones, spotting products a seller has mislabeled or mismapped, and a case tied to their own past work.
The hands-on round gives you a dataset and asks you to build a model. Two candidates report a reranking task, one for smartphone and computer categories. One was allowed to finish the assignment later, and it was converted to a take-home for extra verification.
The final hiring manager round focuses on impact, handling feedback, and why you made specific decisions in past projects. Two candidates were rejected after the process and two received offers, so know your projects well enough to defend each choice.
One candidate with an offer reports a different emphasis: SQL joins, execution order, optimization, third-highest-salary style queries, t-tests, p-values and the central limit theorem. The reports do not say why this process differed. Brush up on SQL and statistics alongside ML.
Synthesized from 4 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Flipkart process.
I found out about the role through Instahyre and got called in for what turned out to be a multi-round process. The first round was an online assessment covering SQL fundamentals and some probability concepts like the central limit theorem. I wasn't entirely prepared for how theoretical it would get, but it helped me understand they were looking for solid foundational knowledge across both practical SQL and statistics.
The technical rounds came next, and SQL was clearly the backbone of the interview. One round focused heavily on joins—I had to write queries joining customer and order tables and handle different join scenarios. Another technical round pushed deeper into execution order in SQL and some moderately complex queries like finding the third highest salary. The interviewer asked me to walk through my thinking on query optimization, which caught me off guard a bit. Beyond SQL, they tested statistical concepts like t-tests and p-values, and asked me to think through machine learning scenarios around random forests and clustering. There was also a case study round where I had to think through feature engineering in the context of recommendation systems, which felt more like real problem-solving than memorized answers.
What stood out was how much they valued communication skills throughout. The interview started with a straightforward introduction about my background and project experience, and they genuinely seemed interested in understanding how I'd approached problems, not just whether I knew the answers. One round included a managerial-style conversation about business understanding—they asked what metrics I'd want to see in a dashboard for a given scenario, which forced me to think beyond just writing correct code.
The whole process took several rounds, but it felt fair and incremental. By the end, I'd covered SQL depth, statistical foundations, machine learning concepts, and soft skills. I received an offer, which made the effort worthwhile. Going in, I underestimated how much they'd emphasize SQL execution order and optimization over just knowing the syntax.
Prep tip from this candidate
Deep dive into SQL execution order and complex multi-table joins with optimization in mind—these were tested repeatedly across rounds. Also prepare for probability and statistics concepts like the central limit theorem, t-tests, and p-values, as they appeared in early screening and technical rounds.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Flipkart
How would you assess if a coin is fair after observing 8 tails in 10 flips?
| Question | |
|---|---|
| Max Quantity | |
| Jars and Coins | |
| Bagging vs Boosting | |
| Same Side Probability | |
| HHT or HTT | |
| The Brackets Problem | |
| Flipping 576 Times | |
| Dijkstra implementation | |
| Underpricing Algorithm | |
| Hurdles In Data Projects | |
| Biased Random Number Generator | |
| Biased five out of six | |
| Possibly Biased Coin | |
| Count Transactions | |
| Bias vs. Variance Tradeoff | |
| All Tails Consecutive | |
| Receipt Scoring Microservices | |
| Merchant Dashboard Design | |
| Walking Robot | |
| Text Editor With OOP | |
| Seller Type Modeling | |
| Why Do You Want to Work With Us | |
| Evaluating Revenue Decline | |
| LRU Cache 1 | |
| Bag of Different Coins | |
| Bias Variance Tradeoff | |
| Meta in an Emerging Market | |
| 2nd Highest Salary | |
| Empty Neighborhoods |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report an ML round covering logistic regression, KNN, linear regression, bias-variance, regularization, ensembles and evaluation metrics. One candidate says derivations and notation were expected, and another says the questions leaned toward deep learning, including transformers and dropout.
Candidates describe a second round built around a product or business scenario. Reported prompts include pricing optimization for mobile phones, detecting mislabeled or mismapped seller products, and a case drawn from the candidate's own experience. Expect to structure the problem end to end.
Candidates report receiving a dataset and building a model, with reranking mentioned by two of them, including smartphone and computer categories. One was given time to submit later, and the task was converted to a take-home. Medium-level DSA questions were also mentioned by one candidate.
Candidates report the final conversation centers on impact analysis, handling feedback, and deep dives into past projects, especially why you made particular decisions. One candidate was rejected after this stage, so it is worth preparing specific project narratives.
One candidate who received an offer describes an online assessment on SQL and probability, then technical rounds on joins, execution order, query optimization, t-tests, p-values, random forests, clustering, and recommendation feature engineering. Treat it as an additional area to prepare.
