Blinkit Data Science Interview Questions & Answers
Curated interview practice questions with answers. These are practice scenarios, not a verified list of past interview questions.
About Blinkit hiring for Data Science roles
Blinkit is an Indian quick commerce company, owned by Eternal (formerly Zomato), that delivers groceries and everyday essentials within minutes. It was earlier known as Grofers and runs a network of dark stores in many cities. It hires software engineers, data analysts, product managers, supply chain and operations professionals to run its fast delivery platform.
For Data Science roles, this Blinkit practice set focuses most on Machine Learning (8), Deep Learning and NLP (6), Statistics and Probability (5), MLOps and Model Deployment (5). It has 2 easy, 34 medium, 4 hard questions, so you can start easy and work up to the harder ones.
Company type
E-commerce & consumer tech
Headquarters
Gurugram, India
Main India offices
Gurugram
Blinkit Data Science interview process
Online coding or analytics assessment
Technical round 1
Technical round 2 (design or case)
Hiring manager round
Typical pattern based on public candidate experiences. Rounds vary by role, team and year.
Solved for BlinkitStart practicing
3 QUESTIONS FREESign up to unlock the full question bank and track your progress.Unlock every question and answer, then continue practising company-wise and package-wise.
✓ 3 questions free • Unlock the rest in seconds
01
What is the difference between layer normalisation and batch normalisation?
Batch normalisation normalises each feature using the mean and variance across the batch, so it depends on batch size and behaves differently in training and inference. Layer normalisation normalises across the features of each single example, so it works with any batch size and with variable-length sequences. That is why transformers use layer norm, often in the pre-norm position, while CNNs traditionally use batch norm.
02
What is the difference between a subquery and a correlated subquery?
A normal subquery runs once on its own and its result is used by the outer query, for example WHERE price > (SELECT AVG(price) FROM products). A correlated subquery refers to columns of the outer query, so it is logically evaluated once per outer row, for example comparing each product with its own category average. Correlated subqueries can be slow; joins or window functions are usually faster alternatives.
03
Write Python code to compute the haversine distance between customer and store locations.
The default 0.5 is rarely right. I would get the cost of each outcome: the value of saving a churner, the cost of a retention offer, and the team's capacity to contact customers. Then I compute expected profit or precision and recall at many thresholds on validation data and pick the one that maximises value within capacity. I would also check that the chosen threshold stays stable over recent months.
05
An analyst tested 20 metrics in one experiment and found one significant. Should you trust it, and how do you correct for this?
Each test at 5 percent significance has a 5 percent false positive chance. Run 20 independent tests and you expect about one false positive even if nothing is real. Corrections include Bonferroni, which divides alpha by the number of tests and is strict, and Benjamini-Hochberg, which controls the false discovery rate and keeps more power. Also reduce the problem by choosing one primary metric before the experiment starts.
06
What are embeddings and vector databases, and why do LLM apps use them?
An embedding model turns text into a vector so that similar meanings are close in space. A vector database stores these vectors with metadata and finds the nearest ones to a query vector quickly using approximate nearest neighbour indexes like HNSW. LLM apps use them for semantic search and retrieval-augmented generation, finding the most relevant documents to put into the prompt. Examples include pgvector, FAISS and dedicated vector stores.
A forest averages many trees trained on bootstrapped rows and random feature subsets, which cuts variance and overfitting at the cost of interpretability.
08
Your RAG chatbot often answers using the wrong documents. How do you fix it?
First check retrieval on its own with a labelled set of questions, since generation cannot fix bad context. Then improve chunking so passages are self-contained, add metadata filters such as product or date, try a better embedding model, and add hybrid search with keywords. A reranker on the top 20 results often helps most. Also rewrite vague user questions into clearer search queries before retrieval.
09
Your sales forecast scores well in k-fold cross-validation but fails in production. What could be wrong with the validation?
Random k-fold mixes future and past rows, so the model trains on data from after the period it predicts. That leaks information and inflates scores. Use time-based splits instead: train on an expanding or rolling window and validate on the next period, like TimeSeriesSplit in scikit-learn. Leave a gap between train and validation when features use lagged values, so the test reflects real forecasting conditions.
10
A key metric dropped 8 percent week over week. How do you check whether it is real?
First rule out data issues: tracking changes, pipeline delays, missing partitions or a changed metric definition. Then compare with the same week last year and look for holidays or seasonality. Check normal weekly variation to see if 8 percent is unusual. Break the drop down by platform, region, app version and user type to find where it is concentrated. Finally, check releases, campaigns or outages that happened that week.
11
Write Python code to compare revenue per user between two groups using Welch's t-test.
It tries candidate splits and picks the one that most reduces impurity — Gini or entropy (information gain) for classification, variance / MSE for regression.
13
Your deployed model gives different predictions than the same model in your notebook. Why?
Usually preprocessing differs between training and serving. Check that the same pipeline object was saved, not just the estimator, and that category order, missing value handling and scaling match. Compare library versions, feature types like strings versus integers, column order, and time zones in date features. Send the same few rows through both paths and compare features step by step to find the first mismatch.
14
How would you design an ML platform for a team of 20 data scientists?
I would start from the team's pain points, then provide shared building blocks: a standard project template, managed notebooks with access to the warehouse, experiment tracking and a model registry, an orchestrator like Airflow for pipelines, a feature store if real-time use cases exist, a standard serving path using containers, and default monitoring. Prefer managed services over custom builds, and add self-serve documentation so teams do not depend on one platform engineer.
15
Your NLP classifier works well on English reviews but fails on Hinglish reviews. What do you do?
First measure the gap with a labelled Hinglish test set. Hinglish mixes scripts and spellings, so tokenisers built for English split it badly. I would try a multilingual or Indic model such as mBERT, XLM-R or IndicBERT, normalise common spellings, and add labelled Hinglish data, possibly with transliteration. Evaluate per language segment and monitor the language mix in production, because it changes over time.
16
What are interaction features and when do they help?
Interaction features combine two or more variables, such as price times quantity, or income divided by number of dependents. They help linear models capture effects where one variable changes the impact of another. Tree models can learn some interactions themselves, but explicit ratios often still help. Too many automatic interactions explode the feature count and overfit, so build them from domain logic and validate each one.
17
Write Python code to run an ADF stationarity test and difference the series if needed.
from statsmodels.tsa.stattools import adfuller
def is_stationary(s, alpha=0.05):
p_value = adfuller(s.dropna())[1]
return p_value < alpha
series = s.copy()
d = 0
while not is_stationary(series) and d < 2:
series = series.diff()
d += 1
print('differencing order d =', d)
18
Your model performance degrades every few months. Would you retrain on a fixed schedule or based on triggers?
Scheduled retraining, like monthly, is simple and predictable but may be too late after a sudden change or wasteful when nothing changed. Trigger-based retraining runs when monitoring shows drift or a performance drop, which reacts faster but needs reliable monitoring. Many teams combine both: a regular schedule plus triggers for big shifts. Either way, a retrained model must pass the same evaluation gate before replacing the current one.
19
Why do residual connections make very deep networks easier to train?
A residual block adds its input to its output, so the layers learn a residual F(x) and the block outputs F(x) plus x. If extra layers are not useful, they can learn to output near zero and the block behaves like identity. The skip path also lets gradients flow directly to early layers, reducing vanishing gradients. This is why ResNets with 50 or 100-plus layers train well where plain deep networks degrade.
20
Write Python code to compute a 95% bootstrap confidence interval for the mean.
Common causes are a learning rate that is too high, exploding gradients, bad input values, and unstable maths like log of zero. I would check the input data for NaN or infinite values, lower the learning rate, add gradient clipping, and make sure the loss uses the stable built-in version with logits. With mixed precision, I would check the loss scaler. Logging gradient norms per step helps find the exact point it breaks.
22
What are primary metrics and guardrail metrics in an experiment?
The primary metric is the one the change is meant to improve and that decides success, such as checkout conversion. Guardrail metrics are ones that must not get worse, such as page load time, refund rate, unsubscribes or revenue per user. Defining both before the test prevents cherry-picking afterwards. A change that lifts the primary metric but breaks a guardrail is usually not shipped without further discussion.
23
Soon after a new model version went live, customer complaints spiked. How do you roll back safely and learn from it?
Roll back immediately to the previous registry version through the deployment pipeline, rather than editing servers by hand, and confirm metrics recover. Then investigate offline: compare predictions of both versions on the same traffic, check feature pipelines and data changes, and review what the offline evaluation missed. Add the failing cases to the test set, and use canary or shadow rollout with alert thresholds for future releases.
24
Write scikit-learn code to get precision, recall and a confusion matrix at a custom threshold of 0.3.
The distribution of sample means approaches a normal distribution as the sample size grows (roughly 30+), whatever the population's shape. It lets us build confidence intervals and run hypothesis tests on means.
26
Pandas: find the top 3 products by revenue in each region.
It checks whether observed category counts differ from expected counts, used for independence between two categorical variables and for goodness of fit.
28
What should you log for a production prediction service?
For each request, log a request ID, timestamp, model version, input features or a hash of them, the prediction and score, and latency. Log errors and fallback use separately. Do not log raw personal data unless needed and allowed; mask it. Aggregate these logs into dashboards for traffic, latency percentiles, error rate and score distribution. Joining logged predictions with later outcomes gives the data for monitoring and retraining.
29
How do you use groupby with multiple aggregations?
The join key is probably not unique on the right side, so each left row matches several right rows and the result multiplies. Check with right['key'].duplicated().sum(). Other causes are null or blank keys matching each other after cleaning, or joining on the wrong column. Fix it by deduplicating or aggregating the right table first, and use merge(..., validate='many_to_one') so Pandas raises an error when the assumption breaks.
Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation.
It predicts from the k nearest labelled points. It is simple but slow at prediction time, sensitive to scaling and irrelevant features, and weak in high dimensions.
Function calling lets an LLM respond with a structured request to call a tool you defined, such as get_order_status with an order_id, instead of plain text. Your code runs the function, returns the result to the model, and the model uses it in its final answer. This connects LLMs to live data, calculators and APIs. You must validate arguments, limit which tools exist and handle errors, because the model can choose wrong inputs.
It lets a model weigh which input positions matter most for each output, computing weighted combinations of values from query-key similarity. It is the core of transformers.
35
Your random forest is accurate but too slow at prediction time for a real-time API. What do you do?
First measure where the time goes. Then reduce trees and maximum depth and check the accuracy loss. Switch to a gradient boosting model, which often needs fewer, smaller trees. Compile the model with tools like ONNX Runtime or Treelite. Cache features or precompute scores for known users. If accuracy allows, a simpler logistic regression may meet the latency budget easily. Always compare options on the same validation set.
Convolution filters slide over the input to detect local patterns, pooling reduces size, and deeper layers combine edges into shapes and objects. Used mainly for images.
37
What is the difference between additive and multiplicative seasonality?
With additive seasonality the seasonal swing is a fixed amount, for example weekends always add about 500 orders regardless of the overall level. With multiplicative seasonality the swing is a percentage of the level, so as the business grows the weekend spike grows too. A quick check is to plot the series: if the peaks widen as the trend rises, use multiplicative, or take a log and model it additively.
38
Your credit model has good overall AUC but performs much worse for applicants from one region. What do you do?
First confirm with per-segment metrics and enough samples, since small groups are noisy. Then check why: fewer training examples from that region, different feature distributions, missing data or labels that reflect past biased decisions. Fixes include collecting more data, reweighting, segment-aware features or a separate calibration. Also review fairness and regulatory requirements, and agree with risk and compliance teams on acceptable performance gaps before launch.
39
Your image classifier trained on 2,000 images memorises the training set within 5 epochs. What do you try?
With so little data, I would use transfer learning from a pretrained model like ResNet or EfficientNet and fine-tune only the top layers first. I would add data augmentation such as flips, crops and colour changes, use dropout and weight decay, and apply early stopping on validation loss. Collecting more labelled images or using a smaller model also helps. I would check for duplicate images across train and validation sets too.
40
Write a Python function to compute IoU between two bounding boxes given as (x1, y1, x2, y2).
Think about quick commerce challenges such as dark store inventory, demand forecasting and delivery time before case discussions.
Practise SQL with time-based analysis and metrics like order volume, delivery time and fill rate for analyst roles.
For engineering roles, revise DSA and design of real-time systems like order tracking and rider assignment.
Keep a one-minute story ready for every project on your CV: problem, your part, tools, result.
Practise the hard questions on this page aloud; Blinkit interviewers usually ask follow-ups on your first answer.
Frequently asked questions
How many rounds are there in the Blinkit Data Science interview?
Candidates usually go through 4 stages: Online coding or analytics assessment; Technical round 1; Technical round 2 (design or case); Hiring manager round. The exact rounds change with the role, team, location and your experience, so confirm the process with the recruiter.
Which topics should I prepare for the Blinkit Data Science interview?
This practice set leans on Machine Learning (8), Deep Learning and NLP (6), Statistics and Probability (5), MLOps and Model Deployment (5). Revise those first, then practise explaining one or two of your own projects end to end, because most Blinkit interviewers ask you to walk through real work.
Are these the exact questions asked at Blinkit?
No. These are curated practice questions that match the skills and difficulty usually tested in e-commerce & consumer tech interviews like Blinkit's. They are not a leaked or verified list of past questions.
Where does Blinkit hire Data Science professionals in India?
Blinkit has large teams in Gurugram. Openings and locations change often, so check the Blinkit careers page and filter by city and experience.