← Uber Interview Insights

Uber·Data Scientist·Technical Phone Screen·Senior

Senior
Jul 2026

Summary

Uber DS interview that was basically one massive experiment design question broken into six parts. Felt like a take-home that someone decided to ask live instead. The depth required was real, this wasn't a 'describe an A/B test' kind of thing.

Questions Asked (6)

Q1

You're running a weekly email campaign to 10 million users. How would you set up the randomization unit and any stratification blocks, and how do you prevent contamination from resends or overlap with other campaigns that same week?

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

I went with user-level randomization pretty quickly but fumbled a bit on the contamination angle.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the goal of the experiment and the constraints of the email campaign, then propose a randomization unit (e.g., user-level) and stratification blocks (e.g., by region or engagement level) to reduce variance. Address contamination by discussing techniques like unique send IDs, deduplication, and coordination with other campaigns to avoid overlap.

Pro tip: Mention that you would use a 'holdout' group to measure long-term effects and that you would monitor for spillover effects via network analysis, especially in a ride-sharing context where user interactions can influence each other.

1. Clarify objectives and constraints

Understand the experiment's goal (e.g., increase engagement, revenue) and constraints like send frequency, legal requirements, and existing campaigns. This informs the randomization and stratification strategy.

2. Choose randomization unit

Select the unit of randomization (e.g., user, device, or email address) based on the experiment's scope and potential interference. For email campaigns, user-level randomization is common to avoid within-user contamination.

3. Define stratification blocks

Identify key covariates (e.g., geography, user activity, past engagement) that correlate with the outcome and create strata to ensure balanced groups. Use stratified randomization to assign users within each block.

4. Prevent contamination from resends

Implement deduplication logic to ensure each user receives only one version of the email. Use unique campaign IDs and suppress users who were already sent the email in the same week.

5. Manage overlap with other campaigns

Coordinate with other teams to avoid overlapping campaigns. Use a centralized system to track user exposure and exclude users from multiple experiments if necessary, or use factorial designs if overlap is unavoidable.

Key Points to Mention

  • Randomization unit: user-level to avoid contamination within user
  • Stratification: by region, user engagement level, or other relevant covariates
  • Deduplication: unique send IDs, suppression lists, and frequency capping
  • Coordination: centralized experiment management system to prevent overlap
  • Holdout groups: to measure long-term effects and detect novelty effects
  • Monitoring: track spillover effects and use intent-to-treat analysis

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

Q2

Walk through the sample size calculation for detecting a +0.18 percentage point absolute lift on a 3% baseline CTR, using 80% power and a two-sided alpha of 0.05. State your assumptions and show the formula.

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

This is where I actually felt okay.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly stating the assumptions: baseline CTR p1=0.03, absolute lift d=0.0018, so p2=0.0318, two-sided alpha=0.05, power=0.80. Then present the standard formula for sample size per variant using the normal approximation, and compute the required sample size. Finally, discuss practical considerations like traffic allocation, duration, and potential pitfalls.

Pro tip: Mention that this calculation assumes independent observations and no clustering, but in practice at Uber you must account for intra-user correlation (e.g., multiple sessions per user) which can inflate required sample size; also consider using sequential testing or CUPED to improve sensitivity.

1. State assumptions and parameters

Clearly define baseline CTR (p1=0.03), absolute lift (d=0.0018), resulting treatment CTR (p2=0.0318), significance level (alpha=0.05, two-sided), and power (1-beta=0.80).

2. Choose the appropriate formula

Use the standard formula for comparing two proportions: n per group = (Z_{1-alpha/2} + Z_{1-beta})^2 * (p1(1-p1) + p2(1-p2)) / d^2, where Z values are from the standard normal distribution.

3. Plug in values and compute

Substitute Z_{0.975}=1.96, Z_{0.80}=0.84, p1=0.03, p2=0.0318, d=0.0018 into the formula and calculate the required sample size per variant.

4. Interpret and adjust for practical considerations

Round up to the nearest integer, then discuss adjustments for unequal allocation, multiple metrics, or clustering (e.g., by user). Also estimate experiment duration based on daily traffic.

Key Points to Mention

  • Baseline CTR and absolute lift must be converted to proportions (0.03 and 0.0018).
  • Two-sided alpha of 0.05 corresponds to Z_{1-alpha/2}=1.96; 80% power corresponds to Z_{1-beta}=0.84.
  • The formula assumes independent observations and normally distributed sample proportions.
  • Sample size is sensitive to the effect size: small lifts require large samples.
  • In practice, consider clustering (e.g., user-level randomization) which may require larger sample sizes.
  • Mention alternative approaches like sequential testing or variance reduction techniques (CUPED) to improve power.

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

Q3

Define a primary success metric and at least two guardrail metrics for this email experiment. Be precise about numerators, denominators, and time windows. Why CTR over open rate?

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

CTR as primary felt right and I could defend it: opens are too easy to inflate with subject line tricks that don't drive real value, clicks are closer to intent.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the experiment's goal (e.g., increasing user engagement with email content) and then define a primary success metric that directly measures that goal, with precise numerator, denominator, and time window. Then propose at least two guardrail metrics that ensure the change doesn't harm other important aspects, such as user experience or long-term retention. Finally, justify why CTR is preferred over open rate by highlighting its closer link to actual user interest and downstream actions.

Pro tip: Acknowledge that open rate can be inflated by Apple's Mail Privacy Protection and other tracking limitations, making CTR a more reliable and actionable metric for driving business outcomes.

1. Clarify Experiment Goal

Restate the experiment's objective to ensure alignment, e.g., 'The goal is to increase user engagement with the email content.'

2. Define Primary Success Metric

Specify a metric like Click-Through Rate (CTR): numerator = unique users who clicked at least one link in the email; denominator = unique users who were sent the email; time window = 7 days from send.

3. Define Guardrail Metrics

Propose at least two guardrails, e.g., Unsubscribe Rate (numerator = unique users who unsubscribed; denominator = unique users who were sent the email; time window = 7 days) and Complaint Rate (numerator = unique users who marked as spam; denominator = unique users who were sent the email; time window = 7 days).

4. Justify CTR over Open Rate

Explain that CTR measures actual engagement and intent, while open rate can be misleading due to privacy features and doesn't guarantee content consumption.

Key Points to Mention

  • Precise numerator and denominator for each metric (e.g., unique users vs. total emails sent).
  • Time window for measurement (e.g., 7 days post-send) to capture delayed actions.
  • CTR reflects active user interest and is less susceptible to inflation from privacy features.
  • Open rate can be inflated by Apple's Mail Privacy Protection and may not indicate true engagement.
  • Guardrail metrics like unsubscribe rate and complaint rate ensure the change doesn't harm user experience.
  • Consideration of downstream metrics (e.g., conversion) to align with business goals.

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

Q4

Leadership wants to peek at results daily and stop the experiment early if there's harm. How do you design a valid sequential monitoring plan that still controls Type I error?

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

I blanked for a second on the exact name of the alpha-spending approach and just described the concept: you pre-commit a spending function across interim looks so the cumulative false positive rate stays at 5%.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging the need to balance early stopping with Type I error control, then propose a sequential testing framework like group sequential designs or alpha spending. Emphasize that the plan must be pre-specified, with clear stopping boundaries and adjusted significance levels, and discuss practical implementation considerations.

Pro tip: Mention that you would simulate the operating characteristics (Type I error, power, expected sample size) under various scenarios to validate the design before deployment, and consider using a Bayesian approach with decision rules if stakeholders need continuous monitoring.

1. Clarify objectives and constraints

Understand what 'harm' means (e.g., negative impact on key metrics) and how frequently leadership wants to peek. Determine the primary metric and acceptable Type I error rate.

2. Choose a sequential monitoring method

Select a method such as group sequential designs with O'Brien-Fleming or Pocock boundaries, or alpha spending functions (e.g., Lan-DeMets) that allow flexible interim looks while controlling overall alpha.

3. Pre-specify the analysis plan

Define the number and timing of interim analyses, stopping rules for harm and futility, and the adjusted alpha at each look. Ensure the plan is documented before the experiment starts.

4. Validate via simulation

Run simulations to estimate Type I error, power, and expected sample size under different scenarios. Adjust boundaries if needed to meet requirements.

5. Implement and monitor

Set up automated monitoring with alerts when boundaries are crossed. Ensure that any deviation from the plan is justified and documented.

Key Points to Mention

  • Type I error inflation from multiple looks and the need for alpha adjustment
  • Group sequential designs and alpha spending functions (e.g., O'Brien-Fleming, Pocock, Lan-DeMets)
  • Stopping boundaries for harm (one-sided) and futility
  • Pre-registration of the analysis plan to avoid p-hacking
  • Simulation to validate operating characteristics
  • Practical considerations: frequency of looks, impact on power, and communication with stakeholders

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

Q5

After 48 hours into the experiment, what diagnostics would you run to catch randomization failures, instrumentation delays, or traffic mix shifts like weekend effects? How do you fix issues without biasing the estimates?

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

Ran through the usual checks: SRM test on arm sizes, covariate balance across engagement tiers and locale, day-of-week distribution check.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by outlining a systematic diagnostic plan that checks sample ratio mismatch (SRM), pre-experiment covariates, and time-based traffic patterns. Then explain how to address issues like instrumentation delays or weekend effects using methods that preserve randomization and avoid peeking biases. Emphasize the importance of pre-registered analysis plans and robust statistical techniques.

Pro tip: Proactively mention that you would set up automated alerts for SRM and other health metrics before the experiment starts, and that any fixes should be applied consistently across all variants to avoid introducing bias.

1. Check Sample Ratio Mismatch (SRM)

Run a chi-squared test to compare the observed assignment ratio to the expected ratio. If p-value is very low, investigate potential causes like logging errors or bot traffic.

2. Validate Instrumentation and Data Pipeline

Compare event counts and timestamps across variants to detect delays or missing data. Check for discrepancies in how metrics are logged between control and treatment.

3. Analyze Traffic Mix and Temporal Patterns

Segment users by day-of-week, hour-of-day, and other dimensions to identify shifts in traffic composition. Use pre-experiment data to establish expected patterns.

4. Diagnose Root Causes and Apply Fixes

If issues are found, determine if they are random or systematic. For systematic issues, apply fixes uniformly across all variants (e.g., reweighting, filtering) and document changes.

5. Ensure Unbiased Estimation

Use methods like CUPED or stratification to adjust for imbalances, and avoid peeking by adhering to pre-registered analysis plans. Consider sequential testing if early stopping is needed.

Key Points to Mention

  • Sample Ratio Mismatch (SRM) detection and its implications
  • Pre-experiment covariate balance checks (e.g., A/A tests)
  • Instrumentation delays and data freshness monitoring
  • Weekend/weekday effects and traffic mix shifts
  • Techniques to avoid bias: CUPED, stratification, consistent fixes across variants
  • Pre-registration and sequential testing to handle peeking

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

Q6

An interim look shows negative CTR lift but higher open rates. Give at least three plausible explanations and explain what you'd do next. Also cover how you'd handle intention-to-treat versus per-protocol analysis and what you'd tell stakeholders.

A/B Testing & ExperimentationRoot Cause AnalysisStakeholder Management
Author's notes

Three causes I gave: subject line drives curiosity opens but the email body doesn't deliver on the promise so nobody clicks; bots or preview panes inflating open counts in treatment; or a rendering issue on certain clients causing click tracking to break.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging the mixed results and framing the need for a careful, multi-faceted analysis. Then, systematically outline plausible explanations, next steps, and how you'd handle ITT vs. PP and stakeholder communication. Emphasize the importance of not overreacting to interim results and considering the broader context.

Pro tip: Mention that interim analyses are often underpowered and that peeking can inflate false positives; suggest pre-registered sequential testing or alpha spending to maintain rigor. This shows statistical maturity and practical experience.

1. Acknowledge and contextualize

Recognize the mixed results and state that interim looks require caution due to multiple testing and limited sample size. Set the stage for a structured investigation.

2. Generate plausible explanations

List at least three explanations: e.g., novelty effect, metric definition mismatch, segment heterogeneity, instrumentation issues, or external factors. Briefly explain each.

3. Outline next steps

Describe actions like checking data quality, segmenting results, running power analysis, waiting for full data, or conducting qualitative research. Prioritize based on impact.

4. Address ITT vs. PP

Explain the difference: ITT includes all randomized users, preserving randomization; PP includes only those who complied. Discuss when each is appropriate and how to handle non-compliance.

5. Communicate with stakeholders

Tailor the message to the audience: emphasize caution, avoid premature conclusions, and propose a plan. Use clear, non-technical language and focus on decision-making.

Key Points to Mention

  • Novelty effect: users may engage more initially, but CTR drops as novelty fades.
  • Metric definition: CTR might be diluted if open rates increase but clicks per open decrease.
  • Segment heterogeneity: treatment effect may vary by user segment (e.g., new vs. existing users).
  • Data quality: check for instrumentation issues, bot traffic, or logging errors.
  • Statistical power: interim analyses are underpowered; avoid peeking without correction.
  • ITT vs. PP: ITT preserves randomization and is conservative; PP can be biased but informative for efficacy.
  • Stakeholder communication: be transparent about uncertainty, avoid overreacting, and propose next steps.

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