Curated interview practice questions with answers. These are practice scenarios, not a verified list of past interview questions.
About Intuit hiring for Data Science roles
Intuit makes financial software including TurboTax, QuickBooks, Credit Karma and Mailchimp. Its Bengaluru development centre works on product engineering, data platforms and AI features for small businesses and consumers. It hires in India for software engineers, quality engineers, data engineers, data scientists and analysts. Teams often work closely with US product groups.
For Data Science roles, this Intuit practice set focuses most on Machine Learning (6), Feature Engineering (6), Generative AI and LLMs (5), MLOps and Model Deployment (5). It has 4 easy, 30 medium, 6 hard questions, so you can start easy and work up to the harder ones.
Company type
Fintech
Headquarters
Mountain View, California, USA
Main India offices
Bengaluru
Intuit Data Science interview process
Online coding assessment
Craft demonstration or take-home exercise
Technical interviews
Hiring manager round
HR round
Typical pattern based on public candidate experiences. Rounds vary by role, team and year.
Solved for IntuitStart 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
How do you use groupby with multiple aggregations?
It tries candidate splits and picks the one that most reduces impurity — Gini or entropy (information gain) for classification, variance / MSE for regression.
03
What is the difference between zero-shot and few-shot prompting?
In zero-shot prompting you give only instructions, like classify this review as positive or negative. In few-shot prompting you also include a few worked examples of input and correct output, which shows the model the format and decision boundary. Few-shot usually improves consistency on custom labels, at the cost of more tokens. Pick examples that cover tricky cases and keep their format identical to what you want back.
It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.
06
What is gradient descent and what does the learning rate do?
An optimisation method that updates weights in the opposite direction of the loss gradient. The learning rate sets the step size: too high and it diverges, too low and it is slow. Variants include SGD, mini-batch and Adam.
07
Write Python code to compute the Population Stability Index for a feature to detect drift.
The model has learned a shortcut: the watermark correlates with the label in your dataset, perhaps because images of one class came from a single source. It will fail when that correlation breaks. Remove or crop the watermarks, rebalance sources across classes, and create a validation set from different sources. Keep using saliency maps and slice-based evaluation to catch other spurious cues such as backgrounds or text.
09
How do you deal with high-cardinality categorical features?
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.
11
Users say your internal chatbot gives outdated policy answers. What is the likely cause and fix?
The knowledge index probably still holds old document versions, or the model is answering from its training data instead of retrieved context. Fix the ingestion pipeline to re-index on document change and remove superseded versions. Store effective dates in metadata and filter or boost the latest. Instruct the model to answer only from provided context and show the source with its date, so users can verify it.
Use the provider's structured output or JSON schema mode when available, which constrains the output to your schema. Otherwise, describe the exact schema in the prompt with an example, and use low temperature. Always parse and validate the response in code, for example with Pydantic, and on failure retry with the error message included. Keep schemas small and flat, and use enums for fields with fixed values.
Bias occurs when the sample does not represent the population. Reduce it with random or stratified sampling, higher response rates and by checking sample demographics against the population.
Three weeks is too little to learn seasonality, so borrow information. Find similar past launches by category, price band and channel, and use their early sales curves as analogues. A global model trained across many products, with product attributes as features, can also forecast a cold-start item. Combine this with planned marketing spend and business judgement, show a wide interval, and update the forecast weekly as real data arrives.
16
A feature won its A/B test with a 4 percent lift, but after full launch the metric barely moved. Why might that happen?
Several reasons are common. Winner's curse: significant results from underpowered tests tend to overestimate the true effect. Novelty effect faded after launch. Seasonality or other launches changed the baseline, so a before-after comparison is misleading. The test population may differ from all users, for example only one platform. Use a holdout after launch to measure real impact, and run tests with adequate power and duration.
Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation.
18
Your model API meets average latency but p99 latency is 900 ms against a 200 ms target. What do you do?
Profile slow requests to see if time goes to feature fetching, preprocessing, the model or the network. Common fixes: precompute and cache features, remove slow database calls from the request path, load the model once at start-up, use a faster runtime like ONNX, and warm up instances. Check garbage collection pauses, cold starts and autoscaling limits. Set timeouts with a safe fallback prediction for rare slow cases.
19
The average delivery time is 30 minutes, yet many customers complain about late orders. How do you investigate?
An average hides the shape of the distribution. Delivery times are usually right-skewed, so a 30-minute mean can include many orders taking 60 minutes or more. Look at the median, the 90th and 95th percentiles and a histogram, then break them down by city, time of day and restaurant. Report the share of orders beyond the promised time, because that is what customers actually experience.
20
What is correlation and why does it not imply causation?
Correlation measures linear association between variables (-1 to 1). A third factor, reverse causality or chance can create it, so causation needs controlled experiments.
L1 (Lasso) adds absolute weights and can shrink some to zero, giving feature selection. L2 (Ridge) adds squared weights and shrinks them smoothly without removing features.
22
Write Pandas code to create lag and rolling features for each customer.
df = df.sort_values(['customer_id', 'month'])
g = df.groupby('customer_id')['spend']
df['spend_lag1'] = g.shift(1)
df['spend_lag3'] = g.shift(3)
# shift first so the current month is not included
df['spend_roll3_mean'] = g.transform(lambda s: s.shift(1).rolling(3).mean())
23
After one-hot encoding you have 2,000 features and training is slow. What do you do?
Find which columns caused the explosion, usually a few high-cardinality categoricals. Group rare categories into Other, or switch those columns to frequency, target or hashing encoding. Use sparse matrices so zeros take no memory. Tree libraries like LightGBM and CatBoost can handle categoricals natively without one-hot. Then remove near-constant columns and check that validation scores hold up with the smaller feature set.
Bias is error from overly simple assumptions (underfitting). Variance is sensitivity to the training data (overfitting). More model complexity lowers bias but raises variance — aim for the lowest total error using cross-validation and regularization.
26
Halfway through a test you discover a bug that broke the treatment for Android users. What do you do?
Stop or pause the affected part, fix the bug and restart the test with fresh data, since mixing broken and fixed periods biases results. Do not simply drop Android users from analysis afterwards, because that can break randomisation balance if the bug affected behaviour or logging. Document the incident, and add pre-launch checks such as running the variant through QA on every platform and monitoring guardrails in the first hours.
27
Write SQL to find the median order value per city.
SELECT city,
PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY amount) AS median_amount
FROM orders
GROUP BY city;
-- PERCENTILE_CONT interpolates; PERCENTILE_DISC returns an actual value
The receptive field is the region of the input image that can influence one activation. Early layers see small patches like edges, and deeper layers see larger regions, enabling recognition of whole objects. It grows with kernel size, depth, stride and pooling, and dilated convolutions increase it without extra parameters. If the receptive field is smaller than the objects you care about, the network cannot capture their full context.
29
Your fraud model's true labels arrive 30 days late. How do you monitor it in production?
Since accuracy cannot be measured quickly, monitor proxies. Track input data drift with PSI per feature, prediction score distribution, the share of transactions flagged, and data quality like missing values. Watch early signals such as chargeback complaints or manual review outcomes on a sample. Set alerts on sudden shifts. When labels arrive, compute real precision and recall by cohort and compare them with the proxy alerts to tune them.
30
What is chunking in RAG and how do you choose a chunk size?
Chunking splits documents into smaller pieces before embedding them, so retrieval returns focused passages that fit in the prompt. Small chunks, like 200 tokens, are precise but lose context; large chunks, like 1,000 tokens, keep context but add noise and cost. Split on natural boundaries such as headings and paragraphs, add some overlap, and keep metadata like title and section. Then tune chunk size using retrieval metrics on real questions.
They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks.
32
Write NumPy code to compute the Dice coefficient between a predicted and a true binary mask.
Target encoding replaces each category with the average target value for that category, for example the churn rate per city. It handles high-cardinality columns compactly. The main risk is leakage: if a row's own target is included in its encoding, the model sees the answer and overfits. Prevent it with out-of-fold encoding, smoothing toward the global mean for rare categories, and fitting only on training data.
34
Pandas: find the top 3 products by revenue in each region.
Pooling downsamples a feature map by summarising small regions. Max pooling keeps the largest value in each window, such as 2x2, and average pooling keeps the mean. It reduces computation, enlarges the receptive field of later layers and gives a little tolerance to small shifts. Many modern networks replace pooling with strided convolutions, but global average pooling before the classifier head is still very common.
36
What is hierarchical forecasting and why is reconciliation needed?
Hierarchical forecasting produces forecasts at several levels, such as SKU, store, city and national totals. If you forecast each level independently, the store forecasts will not add up to the city forecast. Reconciliation adjusts them to be consistent. Bottom-up sums the lowest level, top-down splits the total by proportions, and methods like MinT combine all levels optimally. Consistent numbers matter because finance, supply chain and stores plan from different levels.
37
Write Python code to find the top 3 most similar documents to a query using cosine similarity on embeddings.
Situation: we had to classify 50,000 support tickets into 12 categories with only 600 labelled. Task: reach usable accuracy in a month. Action: I used a pretrained sentence embedding model with logistic regression, then active learning to pick the most uncertain tickets for agents to label in short daily batches. Result: with 2,000 labels we reached 86 percent macro F1. I learned that smart labelling beats waiting for perfect data.
39
Tell me about a time you created documentation that helped your team work with data.
Situation: new analysts kept misusing two similar revenue columns. Task: reduce repeated questions and errors. Action: I wrote a short data dictionary for our main tables, with definitions, known pitfalls and sample queries, and linked it in the team wiki and onboarding checklist. I updated it whenever schemas changed. Result: onboarding time dropped, and the number of revenue mismatches in reports fell noticeably within a quarter.
40
Your model's most important feature is customer_id. What is wrong?
An ID has no real meaning, so high importance means the model is memorising individual customers. It often happens when the same customers appear in both train and test, or when IDs were assigned in time order and act as a proxy for date or signup cohort. I would drop the ID, use a group-based split by customer, and if timing matters, add an explicit feature like account age instead.
Tips to crack the Intuit Data Science interview
Prepare for a craft demonstration, where you may build or present a small working solution and explain your design choices.
Learn how QuickBooks and TurboTax serve small businesses and taxpayers so you can discuss customer problems clearly.
Read about Intuit's AI-driven expert platform strategy and be ready to talk about using data responsibly in financial products.
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; Intuit interviewers usually ask follow-ups on your first answer.
Frequently asked questions
How many rounds are there in the Intuit Data Science interview?
Candidates usually go through 5 stages: Online coding assessment; Craft demonstration or take-home exercise; Technical interviews; Hiring manager round; HR 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 Intuit Data Science interview?
This practice set leans on Machine Learning (6), Feature Engineering (6), Generative AI and LLMs (5), MLOps and Model Deployment (5). Revise those first, then practise explaining one or two of your own projects end to end, because most Intuit interviewers ask you to walk through real work.
Are these the exact questions asked at Intuit?
No. These are curated practice questions that match the skills and difficulty usually tested in fintech interviews like Intuit's. They are not a leaked or verified list of past questions.
Where does Intuit hire Data Science professionals in India?
Intuit has large teams in Bengaluru. Openings and locations change often, so check the Intuit careers page and filter by city and experience.