Data Scientist Interview Questions

Showing 511–540 of 1971 results

• ChimeMedium·Data Scientist
  Write rolling-window SQL over weekly cohorts

  This question evaluates proficiency with SQL window functions and CTEs for time-based cohort aggregation, including rolling ROWS-based sums, LAG-based week-over-week percent change, DENSE_RANK ranking, and handling first- and second-purchase calculations.

  Data Manipulation (SQL/Python)Technical Screen
• CitadelMedium·Data Scientist
  Analyze Correlations and Generate Gaussians

  This question evaluates understanding of correlation properties, feasible dependence constraints in multivariate distributions, construction of joint random variables with specified pairwise correlations, and methods for generating Gaussian samples from uniform random number generators, testing skills in probability, statistics, and simulation within the Machine Learning domain. It is commonly asked to assess both conceptual understanding of dependence structures and positive semidefiniteness and practical application of random variate generation techniques, reflecting a mix of conceptual and applied competencies for data scientist roles.

  Machine LearningTechnical Screen
• CitadelMedium·Data Scientist
  Analyze Town Temperatures and Select Predictors for NYC

  Implement exact pandas and NumPy functions for temperature statistics, single- and two-town regressions, deterministic return values, and greedy five-feature forward selection.

  Machine LearningTechnical Screen
• CitadelHard·Data Scientist
  Build a regression model for wind power output

  This question evaluates competency in feature engineering, physics-informed regression modeling, sensor-data quality handling, model selection and validation, and uncertainty estimation for turbine-level instantaneous power prediction.

  ML System DesignOnline Assessment
• CitadelMedium·Data Scientist
  Calculate Probability of Third Card Being an Ace

  Evaluates conditional probability with cards drawn without replacement after observing at least one Ace. Strong answers define the conditioning event, account for the number of Aces among the first two cards, and compute the probability that the third card is an Ace as 49/825.

  Statistics & MathTechnical Screen
• CitadelEasy·Data Scientist
  Choose Between Precision and Recall for a Classification Decision

  Review precision and recall for imbalanced classification and learn how both change with the decision threshold. Choose an operating point from error costs, capacity, calibration, uncertainty, and segment performance instead of treating either metric as universally best.

  Machine LearningTechnical Screen
• CitadelHard·Data Scientist
  Classify Variable-Length, Unordered Point Sequences with Missing Features

  A staged live-coding ML task: classify whole sequences of 6-feature points as 0 or 1, then adapt the model as lengths vary from 16 to 100, point order stops being meaningful, and some features go missing. Tests model selection, length and permutation invariance, and practical missing-value handling.

  Machine LearningTechnical Screen
• CitadelMedium·Data Scientist
  Compute P(third Ace | Ace in first two)

  This question evaluates understanding of conditional probability, dependence versus independence, symmetry arguments, and combinatorial reasoning in sampling without replacement.

  Statistics & MathTechnical Screen
• CitadelMedium·Data Scientist
  Compute variance of trading profits

  This interview question evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer for Compute variance of trading profits states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathTechnical Screen
• CitadelEasy·Data Scientist
  Decide Whether an LSTM Is Appropriate for Sequence Data

  Decide whether an LSTM is appropriate for sequential data by defining the sequence unit, prediction horizon, masking, and time-safe validation. Compare the model with lag-based and simpler baselines while considering data size, overfitting, latency, and maintainability.

  Machine LearningTechnical Screen
• CitadelMedium·Data Scientist
  Derive Coefficient and Covariance in Regression Analysis

  Evaluates statistics fundamentals across equicorrelation constraints, reverse regression slopes, covariance of uniform order statistics, and change of variables. Strong answers use positive semidefinite matrices, simple-regression identities, order-statistic expectations, and Jacobian transformations.

  Statistics & MathTechnical Screen
• CitadelHard·Data Scientist
  Design Framework for Robust House-Price Prediction Model

  Evaluates robust house-price prediction modeling across diagnostics, features, scale, and validation. Strong answers cover linear regression diagnostics, Cook's distance, leverage, Random Forest complexity controls and variable importance, housing-market feature engineering, leakage prevention, large-scale training, and time-geography validation.

  Machine LearningTechnical Screen
• CitadelHard·Data Scientist
  Design a time-series home-buy decision classifier

  This question evaluates competency in ML system design for time-series decisioning, including rigorous target and horizon definition, temporal feature engineering, time-aware validation, probabilistic calibration and thresholding, leakage controls, model selection trade-offs, and deployment monitoring.

  ML System DesignOnline Assessment
• CitadelHard·Data Scientist
  Design city home-price prediction system

  This question evaluates proficiency in end-to-end machine learning system design for geospatial property price prediction, covering skills such as feature engineering, time- and spatial-aware validation, model selection, interpretability, fairness/compliance, deployment, monitoring, and error analysis.

  Machine LearningTechnical Screen
• CitadelMedium·Data Scientist
  Determine When a Quadratic Has Finite Minimum

  This question evaluates understanding of quadratic optimization, linear algebra, and convexity concepts—particularly properties of quadratic forms, matrix symmetry, and conditions for boundedness in real-valued unconstrained minimization—and why these properties matter for existence of minima.

  Machine LearningTechnical Screen
• CitadelEasy·Data Scientist
  Diagnose and Validate a Regression Model

  Work through regression model validation from split design and leakage prevention to residual analysis and segment-level error checks. The solution explains MAE versus RMSE, multicollinearity, coefficient limits, and when added model complexity is worth the operational cost.

  Statistics & MathTechnical Screen
• CitadelHard·Data Scientist
  Diagnose outliers and influence in linear regression

  This question evaluates a data scientist's competency in diagnosing outliers, high-leverage points, influential observations, and Cook's distance in ordinary least squares regression, including recognition of robust alternatives and rationale for reporting model-review decisions.

  Machine LearningTechnical Screen
• CitadelMedium·Data Scientist
  Discuss PhD coursework and research impact

  This question evaluates a candidate's foundations in empirical modeling, ability to learn from failed approaches, incorporation of feedback, and skill in quantifying research impact within a research or product context.

  Behavioral & LeadershipTechnical Screen
• CitadelHard·Data Scientist
  Estimate OLS via streaming sufficient statistics

  This question evaluates proficiency in streaming/out-of-core linear regression, including computing sufficient statistics with an intercept, assessing numerical stability of normal equations versus QR/SVD or incremental methods, incorporating ridge penalties, and designing parallel fault-tolerant computations.

  Machine LearningTechnical Screen
• CitadelEasy·Data Scientist
  Evaluate Alternative Data for an Investment Pitch

  Evaluate an alternative dataset for an investment pitch through provenance checks, a falsifiable signal hypothesis, and point-in-time validation. Test incremental value, selection bias, revisions, robustness, and costs before presenting the thesis, risks, and invalidation criteria.

  Analytics & ExperimentationTechnical Screen
• CitadelMedium·Data Scientist
  Explain factor leakage checks and IC/ICIR filtering

  This question tests the conceptual understanding of signal quality metrics and the practical steps required to prevent look-ahead bias and overfitting when developing quantitative trading factors.

  Machine LearningTechnical ScreenQuantitative FinanceAlpha ModelingBacktesting
• CitadelMedium·Data Scientist
  Explain multicollinearity and OLS assumptions

  Explain multicollinearity and OLS assumptions evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathTechnical Screen
• CitadelEasy·Data Scientist
  Find constant minimizing absolute error

  This question tests the understanding of L1 loss minimization and the statistical property that the median is the optimal point estimator under absolute error loss.

  Statistics & MathTechnical ScreenStatisticsOptimizationMedian
• CitadelMedium·Data Scientist
  Implement Infinite Fibonacci Generator Using Lazy Evaluation

  This question evaluates understanding of lazy evaluation and Python generator semantics, testing a candidate's ability to work with iterative sequence production and manage state across iterations.

  Coding & AlgorithmsTechnical Screen
• CitadelMedium·Data Scientist
  Implement Piecewise Linear Interpolation and Extrapolation

  Implement piecewise linear interpolation from unsorted knots, including exact-knot behavior and endpoint extrapolation. Practice paired sorting, binary search, and numerical edge cases.

  Coding & AlgorithmsTechnical Screen
• CitadelMedium·Data Scientist
  Implement lazy unique-merge generator for sorted streams

  This question evaluates knowledge of lazy evaluation, Python generators/iterators, merging nondecreasing streams with duplicate elimination, and considerations of time and space complexity in streaming contexts.

  Coding & AlgorithmsTechnical Screen
• CitadelMedium·Data Scientist
  Implement two-pointer unique-pair sum search

  Implement two-pointer unique-pair sum search evaluates algorithm design, data structures, correctness, complexity, edge cases, and implementation details in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Coding & AlgorithmsTechnical Screen
• CitadelMedium·Data Scientist
  Match a string with wildcard pattern recursively

  This problem tests your ability to implement a recursive pattern-matching algorithm that handles specific wildcard characters while managing overlapping subproblems.

  Coding & AlgorithmsTechnical ScreenRecursionDynamic ProgrammingString Matching
• CitadelHard·Data Scientist
  Solve Classic Probability Questions

  This problem tests your ability to calculate the expected stopping time of a stochastic process and your proficiency with the linearity of expectation in the context of sequential probability.

  Statistics & MathOnsiteProbabilityExpectationStochastic Processes
• CitadelHard·Data Scientist
  Solve probability and expectation problems

  This set of problems evaluates probabilistic reasoning and expectation skills across discrete and continuous settings, covering conditional probability with evolving information, convolution and distribution derivation, occupancy/collector problems, symmetry-based variance on manifolds, and sequential selection/optimal stopping.

  Statistics & MathOnsite