Top 50 outliers Interview Questions
The most frequently asked outliers questions across all roles and companies, ranked by real interview frequency. Updated daily.
Explain a practical preprocessing strategy for missing values and outliers before training a supervised learning model.
Publicis Production
PaytmAAutoZoneDesign a leakage-safe preprocessing pipeline for missing and corrupted manufacturing sensor data before LSTM training.
Micron TechnologyClean missing values and outliers in a large pandas dataset while avoiding leakage and preserving useful signal.
Navy Federal Credit UnionImplement an unsupervised anomaly detector for tabular data and evaluate its sensitivity, false-positive rate, and robustness.
KoBold MetalsDesign an interpretable anomaly-detection workflow for geochemical signals while validating against spatial leakage and exploration outcomes.
KoBold MetalsTests judgment in outlier handling and impact on downstream modeling and metrics.
J.B. Hunt Transport
Ormae
KBRNormalize mixed datetime values and flag missing vehicle, mileage, and timestamp fields in Toyota event data.
ToyotaInterpolate missing humidity values from surrounding readings using PostgreSQL window functions and joins.
McKinsey &Clean Salesforce lead data with CTEs, LEFT JOINs, CASE, and COALESCE to standardize missing values for analysis.
SalesforceTests robustness thinking and practical strategies for cleaning and modeling noisy data.
Engage3
Rutgers University
YahooImpute missing values and cap numeric outliers using medians, IQR bounds, and categorical modes.
Product & DesignSign up to see every question
Create a free account to unlock this list and practice real interview questions.