← Meta Interview Insights

Meta·Data Scientist·Technical Phone Screen·Senior

Senior
Jun 2026

Summary

Stat-heavy technical screen for a DS role at Meta. The whole thing revolved around one multi-part experiment design question and they really wanted you to go deep on the math, not just name-drop CUPED.

Questions Asked (3)

Q1

A pre-period covariate explains 36% of the variance in your outcome metric. Walk through how covariate adjustment (e.g. CUPED-style) changes the variance of your treatment effect estimator, and quantify the sample size savings you'd expect.

A/B Testing & ExperimentationTechnical Trade-offs
Author's notes

The 36% number is a gift because it maps directly to variance reduction.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining the CUPED variance reduction formula: with a pre-period covariate explaining ρ² of the variance, the variance of the adjusted estimator is reduced by a factor of (1 - ρ²). Then quantify the sample size savings as the inverse of this factor, i.e., you need (1 - ρ²) times the original sample size, which corresponds to a 1/(1 - ρ²) - 1 reduction in required sample size. Finally, discuss practical considerations like covariate choice, correlation estimation, and potential pitfalls.

Pro tip: Mention that the variance reduction factor (1 - ρ²) assumes the covariate is perfectly measured and the relationship is linear; in practice, you might achieve slightly less reduction due to estimation error in the adjustment coefficient. Also, note that CUPED can be combined with other variance reduction techniques like stratification or regression adjustment for even greater gains.

1. Explain the CUPED adjustment

Describe how CUPED uses a pre-period covariate (e.g., pre-experiment metric) to adjust the outcome metric, typically by subtracting a scaled version of the covariate from the outcome. The scaling factor is chosen to minimize variance, often the covariance between covariate and outcome divided by the variance of the covariate.

2. Derive the variance reduction

Show that the variance of the adjusted estimator is reduced by a factor of (1 - ρ²), where ρ is the correlation between the covariate and the outcome. Since ρ² = 0.36, the variance is reduced by 36%, i.e., the new variance is 64% of the original.

3. Quantify sample size savings

Sample size is proportional to variance, so the required sample size with CUPED is 64% of the original. This means a 36% reduction in sample size, or equivalently, you need 1/(1 - ρ²) = 1/0.64 ≈ 1.5625 times the original sample size without CUPED? Wait, careful: If variance is reduced by 36%, then to achieve the same power, you need 64% of the original sample size. So savings = 36%.

4. Discuss practical implications

Mention that this 36% reduction in sample size can lead to faster experiments, lower costs, or increased power. Also note that the actual savings might be slightly less due to imperfect correlation estimation or if the covariate is not perfectly pre-period.

Key Points to Mention

  • CUPED uses pre-experiment data to reduce variance by adjusting for a covariate.
  • The variance reduction factor is (1 - ρ²), where ρ² is the proportion of variance explained by the covariate.
  • With ρ² = 0.36, the variance is reduced by 36%, leading to a 36% reduction in required sample size.
  • Sample size savings are directly proportional to variance reduction when power is held constant.
  • Practical considerations: covariate must be pre-treatment, correlation estimated from historical data, and adjustment coefficient estimated from the experiment data.
  • CUPED can be combined with other techniques like stratification or regression adjustment for further gains.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

What assumptions does CUPED rely on, and under what conditions would it break down? What diagnostics would you run to check?

A/B Testing & ExperimentationRoot Cause Analysis
Author's notes

I listed the obvious ones: covariate measured pre-treatment, no treatment contamination in the covariate, stable relationship between X and Y across conditions.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining the core assumptions of CUPED: that the pre-experiment covariate is unaffected by treatment and is linearly related to the outcome, and that the variance reduction is correctly estimated. Then discuss conditions where these assumptions break down, such as non-linear relationships, covariate imbalance, or when the covariate is affected by treatment. Finally, outline diagnostics like checking covariate balance, residual plots, and sensitivity analyses to validate the assumptions.

Pro tip: Emphasize that CUPED is not a magic bullet—its effectiveness hinges on the quality and relevance of the pre-experiment covariate. Always validate with a placebo test: run CUPED on a pre-experiment period to ensure it doesn't introduce bias.

1. State the assumptions

Clearly list the key assumptions: (1) the covariate is measured before treatment and is unaffected by it, (2) the relationship between covariate and outcome is linear and stable across treatment groups, and (3) the variance reduction factor is correctly estimated from historical data.

2. Identify breakdown conditions

Discuss scenarios where CUPED fails: non-linear relationships, covariate affected by treatment (e.g., if treatment influences the pre-period metric), heterogeneous treatment effects on the covariate-outcome relationship, and small sample sizes leading to unstable estimates.

3. Propose diagnostics

Suggest diagnostics such as: checking covariate balance between groups, plotting residuals vs. covariate, comparing variance reduction across groups, and running placebo tests on pre-experiment data to detect bias.

4. Discuss mitigation strategies

If assumptions are violated, consider alternatives: using non-linear CUPED (e.g., with splines), stratifying by covariate, or using other variance reduction techniques like stratification or regression adjustment.

Key Points to Mention

  • Covariate must be pre-treatment and unaffected by treatment (exogeneity).
  • Linear relationship between covariate and outcome is assumed; non-linearity can reduce effectiveness.
  • Variance reduction factor (theta) should be estimated from historical data, not the experiment itself.
  • CUPED can introduce bias if the covariate is affected by treatment or if there are time trends.
  • Diagnostics: covariate balance check, residual plots, placebo tests, and sensitivity analysis.
  • Alternatives: non-linear CUPED, stratification, or using multiple covariates.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

Given this setup, would you prefer stratified randomization or post-hoc covariate adjustment, and what's your reasoning?

A/B Testing & ExperimentationTechnical Trade-offs
Author's notes

Said covariate adjustment and they pushed back a bit.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the experimental context and assumptions, then compare stratified randomization and post-hoc covariate adjustment on bias reduction, variance, and operational feasibility. Conclude with a nuanced recommendation that often favors stratified randomization when key covariates are known pre-treatment, but acknowledges post-hoc adjustment as a fallback for unexpected imbalances.

Pro tip: Emphasize that stratification must happen before randomization and requires pre-registered covariates; post-hoc adjustment can introduce bias if not pre-specified, so always discuss the trade-off between design-time control and analysis-time flexibility.

1. Clarify the experimental setup

Ask about the number of covariates, sample size, and whether covariates are known before treatment assignment. This determines the feasibility and necessity of each method.

2. Evaluate bias and variance implications

Explain that stratified randomization balances covariates by design, reducing confounding and increasing precision, while post-hoc adjustment can correct imbalances but may increase variance and risk overfitting.

3. Consider operational and statistical trade-offs

Discuss implementation complexity: stratification requires pre-registration and can be cumbersome with many covariates, whereas post-hoc adjustment is flexible but must be pre-specified to avoid p-hacking.

4. Make a recommendation with caveats

State a preference based on typical scenarios (e.g., prefer stratification for few key covariates) and note that post-hoc adjustment is a valid backup when stratification is impractical.

Key Points to Mention

  • Stratified randomization ensures balance on known covariates, improving power and reducing confounding.
  • Post-hoc covariate adjustment (e.g., CUPED, regression) can increase precision but must be pre-specified to avoid bias.
  • Stratification is limited by the number of covariates and sample size per stratum; too many strata can lead to incomplete randomization.
  • Post-hoc adjustment is useful when unexpected imbalances occur or when covariates are measured post-treatment.
  • Meta's experimentation culture values pre-registration and rigorous design, so stratification is often preferred for key metrics.
  • Always consider the trade-off between design-time control and analysis-time flexibility, and discuss sensitivity analyses.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.