Top 44
Prep plan
Updated weekly · Last refresh Aug 30

Coupang Data Scientist Interview Questions

The questions to prepare for a Coupang Data Scientist interview. Questions from real interview reports rank first. Updated weekly.

44questions
~7htotal time
Track your progressSign up free to work through all 44 questions and resume where you left off.
Start practicing free →
1
SQL & Data ManipulationStart here. 5 questions · ~45 min
2
A/B Testing & Experimentation5 questions · ~45 min
Power Analysis for Survey ExperimentHard

Determine sample size and power for a customer survey or experiment, including MDE, guardrails, and a disciplined decision rule.

MDEPower AnalysisSample SizeCoupang
More A/B Testing & Experimentation questions with a free account

Sign up to see every question

Create a free account to unlock this list and practice real interview questions.

Get my prep plan
3
Metrics4 questions · ~36 min
Diagnose a Conversion DropHard

Investigate whether a conversion drop came from product friction, traffic mix, or an experiment artifact.

Funnel AnalysisConversion RateDiagnosisCoupang
More Metrics questions with a free account
4
Machine Learning5 questions · ~45 min
Handling Missing Data in MLMedium

Explain practical strategies for handling missing data and how to validate that the chosen approach improves model performance.

Feature EngineeringData WranglingSupervised LearningCoupang
More Machine Learning questions with a free account
5
Product Sense4 questions · ~36 min
Identify Top User Pain PointsMedium

Identify the most important user pain points using both qualitative and quantitative data.

User ResearchUser NeedsPain PointsCoupang
More Product Sense questions with a free account
6
System Design3 questions · ~27 min
Monitor Drift in Ad RankingHard

Design monitoring for a large-scale ad ranking system, with feature drift, training-serving skew, and rollback handled as first-class concerns.

Feature StoreFeature DriftModel ServingCoupang
More System Design questions with a free account
7
More topics18 questions · ~162 min
Fraud Detection Batch vs StreamingMedium

Design a fraud pipeline that compares batch, streaming, and hybrid architectures for 120K tx/sec with sub-300 ms decisions and reconciled hourly tables.

Stream ProcessingETLBatch ProcessingCoupang
Statistical vs Practical SignificanceMedium

Explain why a statistically significant experiment result may still be too small to matter for product or business decisions.

Confidence IntervalsExperimentationHypothesis TestingCoupang
More questions with a free account
The finish line: interview-readyComplete all 44 questions to finish this plan.