Swiggy data science candidates report SQL window functions, Python and pandas, ML fundamentals, resume deep-dives, and product problems such as delivery ETA prediction. Reports include full-time and intern interviews.
$32K
Avg. Base Comp
$42K
Avg. Total Comp
2-3 rounds
Typical Rounds
Not reported
Process Length
Swiggy data scientist candidates describe a mix of coding, machine learning theory and business problem-solving, usually anchored in their own projects. Five reports cover both full-time and internship interviews, and four of them ended without an offer.
SQL window functions and aggregations come up repeatedly. One full-time candidate was asked about rank functions, HAVING and GROUP BY. Another had to write a CTE with window functions to find the two highest-paid employees in each department. An intern reported an nth-highest-salary query, and another intern mentioned joins, aggregations and top-performer style questions on delivery data.
Coding is described as light. Candidates mention easy LeetCode-style tasks such as sorting, finding the second largest number or merging sorted arrays. Python questions also covered pandas functions like apply, get_dummies and concat, and one intern wrote NumPy code for assigning points to cluster centroids.
On machine learning, candidates report L1 versus L2 regularization, classifier trade-offs, cross-entropy loss and K-means. They also report linear regression from first principles and Bayes theorem. Resume walkthroughs can be deep. One intern was asked why log transforms were used, why a decision tree beat a random forest, and to derive the 37% OOB figure.
One full-time candidate's final round was problem-solving: features for ETA prediction, improving Blinkit search, and allocating full-time and part-time delivery partners. Another intern also reported applying ML to delivery time, fraud, demand forecasting and recommendations. Prepare to explain your reasoning aloud, since candidates describe collaborative interviewers.
Synthesized from 5 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 Swiggy process.
I recently interviewed for a Data Scientist Internship at Swiggy, and the process focused strongly on practical problem-solving, SQL, Python, machine learning fundamentals, and the ability to apply analytical thinking to real-world business problems.
The interview started with questions about my projects and technical background. I was asked to explain projects such as my Fraud Detection System, Store Demand Forecasting System, and RAG-based Smart Notes Assistant. The interviewer went beyond the resume and asked about the model architecture, preprocessing, evaluation metrics, and why I selected particular approaches.
The next part focused heavily on Python and data handling. I was asked about working with JSON/raw data, Pandas, filtering, grouping, aggregation, handling missing values, and extracting useful information from datasets. The interviewer also tested my ability to translate a business problem into a data-processing solution.
SQL was another important area. Questions involved joins, GROUP BY, aggregations, filtering, and solving problems such as identifying top-performing entities or calculating metrics from transactional/delivery data.
There was also a DSA/problem-solving component, where the emphasis was more on logical thinking and writing an efficient solution rather than extremely difficult competitive-programming problems.
The interview also included ML fundamentals and product-oriented questions. I was asked how machine learning could be applied to Swiggy-specific problems such as delivery-time prediction, fraud detection, demand forecasting, and recommendation systems. This tested not only theoretical knowledge but also my understanding of how models would work in a real product environment.
Overall, the interview was challenging but highly relevant to a Data Scientist role. One of my key learnings was that it is important to explain the thought process clearly and concisely, rather than only knowing the technical answer. The experience helped me understand the importance of combining SQL, Python, ML fundamentals, DSA, and business/product thinking for data-science interviews.
Questions asked: The interview was quite practical and focused on how I approach problems rather than just theoretical definitions. The main areas covered were SQL, Python, machine learning, statistics, DSA, and product/business thinking.
SQL
WHERE and HAVING.GROUP BY, subqueries, and window functions.Python / Data Handling
Machine Learning
The interviewer also asked how ML could be applied to real Swiggy problems. Some examples included:
The important part was not simply giving the ML algorithm. I had to explain the problem definition, data required, features, model selection, evaluation metric, and how I would measure business impact.
The interview also included problem-solving/DSA-style questions, where the interviewer was interested in my approach and complexity rather than just the final answer.
Overall, the interview was a mix of technical fundamentals + practical ML + SQL + product thinking, so I would recommend preparing all four instead of focusing only on machine-learning theory.
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 Swiggy
Write a function can_shift to return whether or not A can be shifted some number of places to get B
| Question | |
|---|---|
| Size of Joins | |
| Z and t-Tests | |
| Forecasting New Year Revenue | |
| Increased Cancellations | |
| Choosing k | |
| Empty Neighborhoods | |
| 2nd Highest Salary | |
| Experiment Validity | |
| Monthly Customer Report | |
| Rolling Bank Transactions | |
| Customer Orders | |
| Comments Histogram | |
| Closest SAT Scores | |
| Subscription Overlap | |
| Top Three Salaries | |
| Upsell Transactions | |
| Button AB Test | |
| First to Six | |
| Merge Sorted Lists | |
| Compute Deviation | |
| Month Over Month | |
| Download Facts | |
| SELECTive Wine Connoisseur | |
| Average Quantity | |
| 500 Cards | |
| Network Experiment Design | |
| Random SQL Sample | |
| Manager Team Sizes | |
| Longest Streak Users |
Synthesized from candidate reports. Individual experiences may vary.
Two full-time candidates report a first round of easy coding plus SQL. Examples include sorting, second-largest number and merging sorted arrays. SQL covered rank functions, HAVING, GROUP BY, CTEs and window functions. One candidate also reports pandas questions. Expect this to be a practical screen rather than hard algorithm work.
Candidates report ML conversations on L1 versus L2 regularization, classifiers, overfitting and cross-entropy loss. One round began with about 15 minutes on a resume project. Interns report resume-driven probing of preprocessing, metrics, K-means, linear regression and Bayes theorem, so you should be ready to justify each modeling choice.
Intern reports describe applied Python, NumPy, pandas and SQL tasks. These include centroid assignment for a clustering setup, nth-highest salary queries, data cleaning, grouping and aggregation. One intern said the DSA portion focused on logical, efficient solutions rather than very hard competitive-programming problems.
One on-campus candidate describes a third problem-solving and managerial round on ETA features, Blinkit search improvements and delivery-partner allocation. Another candidate describes the third round as HR and behavioral. One intern also reported product-oriented ML questions, so prepare for both business reasoning and teamwork discussion.