← Netflix Interview Insights

Netflix·Data Scientist·Technical Phone Screen·Senior

Senior
May 2026

Summary

Netflix data scientist screen, heavy on stats and experiment design fundamentals. Six questions back to back, no coding, just pure probability and methodology. Felt more like a grad school oral exam than a job interview.

Questions Asked (6)

Q1

State Bayes' rule and apply it: given a disease prevalence of 1%, a test sensitivity of 99%, and a false positive rate of 5%, what is the probability of actually having the disease given a positive test result?

Product Analytics & Metrics
Author's notes

I knew the formula but the number still surprised me when I computed it.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly stating Bayes' rule and defining each term in the context of the problem. Then plug in the given values (prevalence, sensitivity, false positive rate) and compute the posterior probability step by step, explaining the intuition behind the result.

Pro tip: Emphasize that the low prevalence (1%) makes the base rate dominate, so even with a highly sensitive test, the probability of having the disease given a positive result is surprisingly low. This demonstrates an understanding of the base rate fallacy, which is crucial in data science and product analytics.

1. State Bayes' rule

Write the formula: P(D|+) = P(+|D) * P(D) / P(+), and define each term: P(D) is prevalence, P(+|D) is sensitivity, P(+|¬D) is false positive rate.

2. Identify given values

Assign the numbers: P(D) = 0.01, P(+|D) = 0.99, P(+|¬D) = 0.05. Note that P(¬D) = 0.99.

3. Compute total probability of a positive test

Calculate P(+) = P(+|D)*P(D) + P(+|¬D)*P(¬D) = 0.99*0.01 + 0.05*0.99 = 0.0099 + 0.0495 = 0.0594.

4. Apply Bayes' rule

Compute P(D|+) = (0.99 * 0.01) / 0.0594 ≈ 0.1667, or about 16.7%.

5. Interpret the result

Explain that despite a positive test, the probability of having the disease is only ~16.7% due to the low base rate, illustrating the base rate fallacy.

Key Points to Mention

  • Bayes' rule formula and its components
  • Definition of prevalence, sensitivity, and false positive rate
  • Calculation of the total probability of a positive test
  • Final posterior probability (~16.7%)
  • Base rate fallacy and its implications
  • Real-world relevance in data science and product analytics (e.g., A/B testing, anomaly detection)

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

Q2

In an observational study, why does adding control variables sometimes change the coefficient on the main variable you care about? And when can adding controls actually make things worse?

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

Talked through omitted variable bias fine, but fumbled a bit on the second part.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining that adding controls can change the main coefficient due to confounding or mediation, then discuss how controls can introduce bias through collider stratification or overcontrol. Emphasize the importance of using causal diagrams (DAGs) to decide which variables to control for, and highlight the trade-off between bias and variance.

Pro tip: Mention that in observational studies, the goal is to estimate a causal effect, so you must control for confounders but avoid controlling for mediators or colliders. At Netflix, where experimentation is gold standard, you'd validate observational findings with A/B tests when possible.

1. Explain why coefficients change

Adding controls can change the main coefficient because it adjusts for confounding (removing omitted variable bias) or because it blocks a causal pathway (mediation), altering the interpretation.

2. Distinguish confounders vs. mediators vs. colliders

Confounders should be controlled to get unbiased estimates; mediators should not be controlled if you want the total effect; colliders should never be controlled as they induce spurious associations.

3. Discuss when adding controls makes things worse

Controls can introduce bias (e.g., collider bias, overcontrol bias), increase variance, or reduce statistical power, especially with small samples or many controls.

4. Recommend using causal diagrams (DAGs)

Use DAGs to identify the minimal sufficient adjustment set and avoid controlling for inappropriate variables.

5. Connect to experimentation

Acknowledge that observational findings should be validated with randomized experiments (A/B tests) when possible, as they eliminate confounding.

Key Points to Mention

  • Omitted variable bias and confounding
  • Mediation and overcontrol bias
  • Collider bias (Berkson's paradox)
  • Causal diagrams (DAGs) for variable selection
  • Bias-variance trade-off when adding controls
  • Importance of randomization in A/B testing to avoid these issues

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

Q3

State the Central Limit Theorem and explain what it means practically for the sampling distribution of the sample mean.

Product Analytics & Metrics
Author's notes

Straightforward.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by formally stating the Central Limit Theorem (CLT), then immediately pivot to its practical implications for the sampling distribution of the sample mean. Use a concrete example relevant to Netflix (e.g., average watch time per user) to illustrate how the CLT enables inference and hypothesis testing even when the population distribution is unknown.

Pro tip: Emphasize that the CLT applies to the sampling distribution of the mean, not the raw data, and that the approximation improves with sample size—this shows you understand the subtlety and can avoid common misinterpretations.

1. State the CLT formally

Define the Central Limit Theorem: For a population with finite mean μ and finite variance σ², the sampling distribution of the sample mean approaches a normal distribution with mean μ and variance σ²/n as the sample size n becomes large, regardless of the population's shape.

2. Explain the practical meaning

Describe how the CLT allows us to use normal-based methods (e.g., confidence intervals, hypothesis tests) for the sample mean even when the population distribution is not normal, provided n is sufficiently large.

3. Discuss the role of sample size

Mention that the required n depends on the population's skewness and kurtosis; for highly skewed data, larger n is needed. Typically n ≥ 30 is a rule of thumb, but it's not universal.

4. Connect to Netflix context

Give a Netflix-specific example: e.g., estimating average daily watch time per user. Even if individual watch times are skewed, with a large enough sample, the sample mean will be approximately normal, enabling A/B testing and metric comparisons.

5. Highlight implications for inference

Summarize that the CLT underpins many statistical procedures used in product analytics, such as t-tests, confidence intervals for means, and control charts, making it foundational for data-driven decision-making.

Key Points to Mention

  • The CLT applies to the sampling distribution of the sample mean, not the distribution of individual observations.
  • The mean of the sampling distribution equals the population mean (μ), and its standard deviation is σ/√n (standard error).
  • The approximation becomes more accurate as sample size increases, but the rate depends on the population distribution.
  • The CLT holds for independent and identically distributed (i.i.d.) random variables with finite variance.
  • Practical use: enables construction of confidence intervals and hypothesis tests for means without assuming normality of the population.
  • Common misconception: the CLT does not make the raw data normal; it only makes the sampling distribution of the mean approximately normal.

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

Q4

For X drawn from a uniform distribution on the interval (a, b), derive the mean and variance.

Product Analytics & Metrics
Author's notes

Easy.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly stating the probability density function (PDF) of the uniform distribution on (a, b). Then derive the mean using the integral of x times the PDF, and the variance using the definition Var(X) = E[X^2] - (E[X])^2. Show each step of the integration and simplify to the final formulas.

Pro tip: Emphasize that the derivation is straightforward but be prepared to explain why the variance formula makes intuitive sense (e.g., it scales with the square of the interval width). Also, mention that in practice, you might use these formulas to quickly compute summary statistics for uniformly distributed data.

1. State the PDF

Write the probability density function of X: f(x) = 1/(b-a) for a < x < b, and 0 otherwise.

2. Derive the mean

Compute E[X] = ∫_a^b x * f(x) dx = (1/(b-a)) ∫_a^b x dx = (a+b)/2.

3. Derive E[X^2]

Compute E[X^2] = ∫_a^b x^2 * f(x) dx = (1/(b-a)) ∫_a^b x^2 dx = (a^2 + ab + b^2)/3.

4. Compute variance

Use Var(X) = E[X^2] - (E[X])^2 = (a^2 + ab + b^2)/3 - ((a+b)/2)^2 = (b-a)^2/12.

5. Summarize and interpret

State the final results: mean = (a+b)/2, variance = (b-a)^2/12. Optionally, discuss the standard deviation and its interpretation.

Key Points to Mention

  • The PDF of a continuous uniform distribution is constant over the interval.
  • Integration is used to compute expected values for continuous random variables.
  • The mean is the midpoint of the interval, which is intuitive.
  • The variance depends only on the width of the interval (b-a), not on the location.
  • The derivation uses the identity Var(X) = E[X^2] - (E[X])^2.
  • The result can be generalized to any uniform distribution, and is useful in simulations and random number generation.

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

Q5

Write out the one-sample t-statistic for testing whether the population mean equals some hypothesized value, given a sample mean, sample standard deviation, and sample size.

A/B Testing & Experimentation
Author's notes

No issues here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

State the formula for the one-sample t-statistic clearly: t = (x̄ - μ₀) / (s / √n), defining each term. Then briefly explain the intuition: it measures how many standard errors the sample mean is from the hypothesized mean, and note the t-distribution with n-1 degrees of freedom.

Pro tip: Mention that in A/B testing contexts, this is essentially the same as comparing a treatment mean to a baseline, and that using the sample standard deviation (not population) is why we use t rather than z.

1. Define the terms

Identify x̄ as the sample mean, μ₀ as the hypothesized population mean, s as the sample standard deviation, and n as the sample size.

2. Write the formula

State the t-statistic: t = (x̄ - μ₀) / (s / √n). Emphasize that the denominator is the standard error of the mean.

3. Explain the intuition

Describe the t-statistic as the number of standard errors the sample mean is away from the hypothesized mean, indicating how unusual the sample would be if the null hypothesis were true.

4. Mention the distribution

Note that under the null hypothesis, the t-statistic follows a t-distribution with n-1 degrees of freedom, which accounts for the uncertainty in estimating the standard deviation.

Key Points to Mention

  • Formula: t = (x̄ - μ₀) / (s / √n)
  • Sample standard deviation s is used, not population σ
  • Standard error of the mean is s / √n
  • Degrees of freedom = n - 1
  • The t-statistic measures the distance in standard errors
  • Applicable in A/B testing for comparing a sample mean to a baseline

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

Q6

Define effect size and Minimum Detectable Effect. How do statistical power, variance, sample size, and significance threshold each affect the MDE?

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

This one tripped me up more than it should have.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly defining effect size and MDE, then explain how statistical power, variance, sample size, and significance threshold each influence MDE. Use a formula or relationship to show the interplay, and conclude with practical implications for experiment design at Netflix.

Pro tip: Emphasize that MDE is a design parameter, not a post-hoc calculation—it should be determined before the experiment to ensure adequate power. Also, mention that reducing variance (e.g., via stratification or CUPED) can lower MDE without increasing sample size.

1. Define Effect Size and MDE

Effect size quantifies the magnitude of a difference (e.g., relative lift, Cohen's d). MDE is the smallest true effect an experiment can detect with a given power and significance level.

2. Explain the MDE Formula

Present the relationship: MDE ∝ (Z_{1-α/2} + Z_{1-β}) * σ / √n, where α is significance threshold, β is Type II error, σ is variance, and n is sample size per group.

3. Analyze Each Factor's Impact

Describe how increasing power (1-β) or significance threshold (α) increases MDE; increasing variance increases MDE; increasing sample size decreases MDE.

4. Discuss Practical Implications

Explain how to use these relationships to design experiments: e.g., to detect smaller effects, increase sample size, reduce variance, or accept lower power.

Key Points to Mention

  • Effect size can be absolute or relative, and MDE is often expressed as a relative lift.
  • Statistical power (1-β) is typically set at 80% or 90%; higher power requires larger sample size or larger MDE.
  • Significance threshold (α) is usually 0.05; stricter thresholds (smaller α) increase MDE.
  • Variance (σ²) directly affects MDE: higher variance requires larger sample size to detect the same effect.
  • Sample size (n) has an inverse square root relationship with MDE: quadrupling n halves MDE.
  • MDE is a function of the test's design parameters and should be determined before running the experiment.

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