Interview guide
Data Scientist interview questions
Data science interviews combine statistics, machine learning, coding (Python/SQL) and product sense. The strongest candidates connect every model to a business decision.
Role-specific questions
1.Tell me about a model you built that changed a decision.
Why they ask: Impact over technique.
How to answer: The business question, the baseline, your approach, how you validated it, and the decision or metric that changed.
2.How would you evaluate a classifier for fraud detection?
Why they ask: Metric choice with imbalanced data.
How to answer: Accuracy is misleading; use precision/recall, PR-AUC and the cost of each error type. Pick a threshold with the business.
3.How do you design an A/B test?
Why they ask: Experimentation is core in product data science.
How to answer: Hypothesis, primary metric and guardrails, sample size from minimum detectable effect, randomisation unit, run length, and how you’d avoid peeking.
4.Your model performs well offline but poorly in production. Why?
Why they ask: Real-world ML experience.
How to answer: Data drift, leakage in training, train/serve skew, feedback loops. Explain how you’d monitor and diagnose.
5.Explain a complex result to a non-technical stakeholder.
Why they ask: Communication decides whether your work is used.
How to answer: Lead with the recommendation, one chart, the uncertainty in plain words, and what you need from them.
6.Write a SQL query to find each user’s first purchase.
Why they ask: SQL is used daily.
How to answer: Window functions (ROW_NUMBER over user ordered by time) or a MIN subquery join. Talk through ties and nulls.
Reading answers isn’t practising them
Get questions built from your résumé and a real job post, answer out loud, and get STAR scoring, follow-up questions and pace feedback. Free to start.
Behavioral questions
Every data scientist interview includes these. Answer with STAR: the Situation, your Task, the Actions you took, and the Result — with a number where you can.
7.Tell me about yourself.
Why they ask: It sets the frame for the whole interview and tests whether you can summarise your career around what they need.
How to answer: 90 seconds: present (current role and one proud result), past (the path that got you here), future (why this role is the logical next step). Skip your life story.
8.Tell me about a time you disagreed with a teammate or manager.
Why they ask: They want to see that you can push back with evidence and still commit once a decision is made.
How to answer: Use STAR. Show how you understood their view, what data you brought, how it was resolved — and what you did if you lost the argument.
9.Describe a project that failed or didn’t go to plan.
Why they ask: Ownership and learning matter more than a perfect track record.
How to answer: Pick a real failure you contributed to. Spend most of the answer on what you changed afterwards — ideally with a later result that proves it.
10.Tell me about a time you had to deliver under a tight deadline.
Why they ask: They’re testing prioritisation and communication, not heroics.
How to answer: Explain what you cut or re-scoped, who you told and when, and the outcome. Mention the trade-off you consciously accepted.
11.What’s an accomplishment you’re most proud of?
Why they ask: It shows what you value and the scale you operate at.
How to answer: Choose one relevant to this role. Make your personal contribution clear (“I”, not “we”) and end with a number.
12.Why do you want to work here?
Why they ask: They want evidence you chose them — not just any job.
How to answer: Connect something specific (their product, a recent launch, how the team works) to your experience and what you want to do next.