← Amazon Interview Insights

Amazon·Data Scientist·Technical Phone Screen·Senior

Senior
Jun 2026

Summary

Amazon DS interview that was essentially one long A/B testing deep dive. They really do want you to know the full lifecycle, not just the stats part. Felt like a technical phone screen but the depth they expected was closer to onsite level.

Questions Asked (7)

Q1

Walk me through how you would design, run, and analyze an A/B test end to end for a product change.

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

This is the kind of question that sounds manageable until you're 10 minutes in and realize you've only covered hypothesis framing and the interviewer is still waiting.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Structure your answer around a clear experimental lifecycle: define the hypothesis and metrics, design the test with proper randomization and power analysis, execute while monitoring for validity threats, and analyze results with statistical rigor and business context. Emphasize how you ensure trustworthy results and translate findings into actionable product decisions.

Pro tip: At Amazon, always tie your experiment to a primary success metric and guardrail metrics, and be prepared to discuss how you'd handle practical challenges like low sample size, novelty effects, or multiple testing corrections.

1. Define Hypothesis and Metrics

Clearly state the product change, the expected impact, and the primary metric (e.g., conversion rate) plus guardrail metrics (e.g., latency, revenue). Ensure metrics are aligned with business goals and measurable.

2. Design the Experiment

Determine randomization unit (e.g., user, session), sample size via power analysis (considering baseline rate, MDE, alpha, power), and duration to capture full weekly cycles. Plan for control/treatment split and any stratification.

3. Run and Monitor the Test

Launch the experiment, monitor for data quality issues (e.g., sample ratio mismatch), and avoid peeking or early stopping unless using sequential testing. Ensure consistent implementation and track any external factors.

4. Analyze Results

Perform statistical tests (e.g., t-test, chi-square) on primary and guardrail metrics, check for practical significance, and adjust for multiple comparisons if needed. Segment analysis can reveal heterogeneous effects.

5. Decide and Communicate

Interpret results in business context, recommend ship/no-ship, and document learnings. Consider follow-up experiments or iterative improvements based on findings.

Key Points to Mention

  • Hypothesis formulation and metric selection (primary vs. guardrail)
  • Power analysis and sample size calculation
  • Randomization and avoiding common pitfalls (e.g., SRM, novelty effects)
  • Statistical significance vs. practical significance
  • Multiple testing corrections (e.g., Bonferroni, FDR)
  • Business impact and decision-making based on results

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

Q2

How do you choose the unit of randomization and define what counts as an exposed user?

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

Fumbled this a bit.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining that the choice of randomization unit and exposure definition should align with the experiment's goal, the treatment mechanism, and the metric of interest. Then walk through a structured framework that considers interference, dilution, and practical constraints. Emphasize that the decision is a trade-off between statistical validity and operational feasibility.

Pro tip: Always define exposure based on when the treatment actually takes effect, not just when the user is assigned. This avoids dilution and ensures you measure the true causal effect.

1. Clarify the experimental goal and metric

Identify the primary metric and the causal question. Determine whether the treatment affects individual users or groups (e.g., households, sessions).

2. Assess interference and spillover

Evaluate if users interact with each other or share resources. If interference is likely, consider cluster randomization (e.g., by household, geographic region) to avoid contamination.

3. Choose the randomization unit

Select the unit that minimizes interference while maximizing statistical power. Common units: user, session, page view, or cluster. Document the rationale.

4. Define exposed users precisely

Specify the trigger that indicates a user actually received the treatment (e.g., viewed the feature, completed a qualifying action). Exclude users who were assigned but never exposed to avoid dilution.

5. Validate and iterate

Check for sample ratio mismatch, ensure exposure definition is measurable and consistent, and run a holdback or A/A test to validate the setup.

Key Points to Mention

  • Randomization unit options: user, session, page view, cluster (e.g., household, geographic region).
  • Interference/spillover effects and how they influence unit choice (e.g., social networks, shared inventory).
  • Exposure definition: intent-to-treat vs. treatment-on-treated, and the importance of avoiding dilution.
  • Statistical power and sample size implications of different randomization units.
  • Practical constraints: data availability, tracking feasibility, and business rules.
  • Amazon-specific examples: Prime eligibility, Alexa device usage, or marketplace seller experiments.

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

Q3

How do you determine sample size, minimum detectable effect, and statistical power for your experiment?

A/B Testing & Experimentation
Author's notes

Went fine.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the problem: define the metric, then explain how you balance statistical power, significance level, and minimum detectable effect (MDE) to compute sample size. Emphasize that the choice of MDE is a business decision, and that you often iterate between MDE and sample size to align with practical constraints.

Pro tip: At Amazon, always tie your sample size calculation to business impact—e.g., 'We need to detect a 1% lift in conversion because that translates to $X million in revenue.' This shows you think like an owner, not just a statistician.

1. Define the metric and hypothesis

Clearly state the primary metric (e.g., conversion rate, revenue per user) and the null and alternative hypotheses. Ensure the metric is sensitive to the change and aligns with business goals.

2. Set statistical parameters

Choose significance level (α, typically 0.05) and power (1-β, typically 0.8). These are standard but can be adjusted based on risk tolerance and multiple testing corrections.

3. Determine the minimum detectable effect (MDE)

Work with stakeholders to define the smallest effect size that is practically meaningful. This is often a business decision based on cost-benefit analysis, not just statistical convenience.

4. Calculate sample size

Use the formula or simulation to compute required sample size per variant, given α, power, MDE, and baseline metric variance. For proportions, use the standard formula; for continuous metrics, use t-tests or simulations.

5. Validate and iterate

Check if the required sample size is feasible given traffic and duration. If not, adjust MDE, power, or consider sequential testing. Always document assumptions and trade-offs.

Key Points to Mention

  • Trade-offs between power, significance level, and MDE: increasing power or reducing MDE requires larger sample size.
  • Business context: MDE should be tied to ROI or strategic importance, not just statistical significance.
  • Use of historical data or pilot studies to estimate baseline variance and conversion rates.
  • Consideration of multiple testing corrections (e.g., Bonferroni) when multiple metrics or variants are involved.
  • Practical constraints: traffic volume, experiment duration, and novelty effects.
  • Alternative approaches like sequential testing or Bayesian methods when fixed-horizon sample size is impractical.

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

Q4

What data quality checks do you run before trusting your experiment results?

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

SRM was the first thing I mentioned and I could tell that landed well.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Structure your answer around a systematic validation pipeline that covers pre-experiment, in-experiment, and post-experiment checks. Emphasize how you automate these checks to catch issues early and ensure trustworthy results, and tie your approach to business impact.

Pro tip: Mention that you always validate your experiment infrastructure with an A/A test before running an A/B test, and that you monitor data quality metrics in real-time to catch issues like sample ratio mismatch immediately.

1. Pre-experiment validation

Verify that the experiment setup is correct by checking randomization, sample size, and metric definitions. Run an A/A test to ensure there's no underlying bias or technical issues.

2. In-experiment monitoring

Continuously monitor data quality metrics such as sample ratio mismatch (SRM), missing data, and outliers. Set up alerts for anomalies to address issues in real-time.

3. Post-experiment data quality checks

After the experiment concludes, perform thorough checks on data completeness, consistency, and distribution. Validate that the data matches expectations and investigate any discrepancies.

4. Statistical validity checks

Ensure that statistical assumptions are met, such as normality, independence, and homogeneity of variance. Check for novelty effects, primacy effects, and other biases that could skew results.

5. Automation and documentation

Automate the data quality checks to run consistently and document the process. This ensures reproducibility and allows for quick identification of issues in future experiments.

Key Points to Mention

  • Sample Ratio Mismatch (SRM) detection and its implications
  • A/A tests to validate experiment infrastructure
  • Data completeness and consistency checks (e.g., missing values, duplicates)
  • Outlier detection and handling
  • Statistical power analysis and sample size validation
  • Automated monitoring and alerting for data quality issues

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

Q5

How do you handle the multiple testing problem and what's your view on sequential peeking at results?

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

Honestly the part I felt shakiest on.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging the multiple testing problem and its impact on false positive rates, then explain common correction methods like Bonferroni or FDR. Discuss sequential peeking as a trade-off between early detection and inflated Type I error, and mention sequential testing methods that allow valid early stopping. Conclude with a practical recommendation for Amazon's experimentation culture, emphasizing rigor and customer impact.

Pro tip: Show that you understand the business context: at Amazon, speed and customer obsession matter, so propose a balanced approach using sequential testing to enable early decisions without sacrificing statistical validity. Mention that you'd pre-register the analysis plan and use tools like sequential p-values or alpha spending.

1. Define the problem

Explain what multiple testing is and why it increases false positives. Mention that peeking at results repeatedly is a form of multiple testing.

2. Classic corrections

Describe standard methods like Bonferroni, Holm-Bonferroni, or Benjamini-Hochberg FDR, and their trade-offs between power and error control.

3. Sequential testing approaches

Introduce sequential testing methods such as group sequential designs, alpha spending functions, or always-valid p-values that allow valid interim analyses.

4. Practical recommendation

Propose a strategy: pre-register the analysis plan, use sequential testing if early stopping is desired, and apply corrections when testing multiple metrics or variants.

5. Amazon context

Tie it to Amazon's culture: emphasize customer impact, speed, and data-driven decisions, and mention how you'd balance rigor with agility.

Key Points to Mention

  • Family-wise error rate (FWER) and false discovery rate (FDR)
  • Bonferroni correction and its conservativeness
  • Sequential probability ratio test (SPRT) or group sequential designs
  • Alpha spending functions (e.g., O'Brien-Fleming, Pocock)
  • Always-valid p-values or Bayesian approaches
  • Pre-registration of analysis plan to avoid p-hacking

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

Q6

How do you decide whether to launch based on experiment results? How do you weigh statistical significance against practical significance?

A/B Testing & ExperimentationProduct StrategyCross-functional Alignment
Author's notes

This is where the conversation got more interesting.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining that you evaluate both statistical and practical significance, but the decision to launch depends on the business context and expected impact. Emphasize that you consider the confidence interval, effect size, and potential risks, and align with stakeholders on the decision criteria before the experiment.

Pro tip: Frame the decision in terms of expected value and risk: even if results are statistically significant, if the practical impact is negligible or negative for key segments, it's not worth launching. Always tie back to the original hypothesis and business metrics.

1. Define success criteria upfront

Before the experiment, establish the primary metric, minimum detectable effect (MDE), and the threshold for practical significance with stakeholders. This ensures alignment and prevents post-hoc rationalization.

2. Assess statistical significance

Check if the observed effect is statistically significant (e.g., p-value < 0.05) and examine the confidence interval to understand the range of plausible effects. Consider power and sample size to avoid false negatives.

3. Evaluate practical significance

Determine if the effect size is large enough to matter for the business. Consider the cost of implementation, potential revenue impact, and whether the lift justifies the investment. Use domain knowledge to judge if the effect is meaningful.

4. Consider secondary metrics and guardrails

Look at secondary metrics and guardrail metrics to ensure no negative impact on user experience or other key areas. A statistically significant win on the primary metric might be offset by declines elsewhere.

5. Make a decision and communicate

Weigh all evidence and make a recommendation: launch, iterate, or abandon. Clearly communicate the rationale, including trade-offs, and align with cross-functional partners on next steps.

Key Points to Mention

  • Statistical significance (p-value, confidence intervals) vs. practical significance (effect size, business impact)
  • Minimum Detectable Effect (MDE) and power analysis
  • Cost-benefit analysis and expected value
  • Guardrail metrics and potential negative impacts
  • Alignment with stakeholders on decision criteria before the experiment
  • Iterative approach: if not significant, consider whether to extend, refine, or stop

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

Q7

What are the most common pitfalls in A/B testing, particularly around interference, noncompliance, and missing data?

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

Network effects was the one I spent the most time on since Amazon's scale makes spillover a real concern.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining each pitfall clearly and explaining why it threatens causal inference in A/B tests. Then, for each pitfall, describe how it manifests in practice and propose detection and mitigation strategies. Finally, tie your answer back to Amazon's scale and customer obsession by emphasizing rigorous experimentation and guardrail metrics.

Pro tip: Show you understand the trade-offs: e.g., fixing interference often requires cluster randomization, which reduces power; addressing noncompliance may need instrumental variables or CACE analysis. Mentioning these nuances signals deep expertise.

1. Define the pitfalls

Briefly define interference, noncompliance, and missing data, and explain how each violates assumptions of standard A/B testing (e.g., SUTVA, randomization, missingness at random).

2. Explain impact on causal inference

Describe how each pitfall biases effect estimates: interference can dilute or amplify effects, noncompliance breaks randomization, and missing data can introduce selection bias.

3. Detection strategies

Outline methods to detect each issue: e.g., checking for spillover via network analysis, measuring compliance rates, and analyzing missing data patterns (e.g., Little's MCAR test).

4. Mitigation techniques

Propose solutions: cluster randomization or switchback tests for interference, intent-to-treat or instrumental variables for noncompliance, and imputation or weighting for missing data.

5. Amazon-specific considerations

Connect to Amazon's context: large-scale experiments, customer behavior interdependence, and the need for robust guardrail metrics to protect customer experience.

Key Points to Mention

  • SUTVA violation and interference (e.g., social networks, marketplaces)
  • Noncompliance: one-sided vs. two-sided, dilution of effect, ITT vs. PP analysis
  • Missing data mechanisms: MCAR, MAR, MNAR, and their implications
  • Cluster randomization and switchback experiments as solutions
  • Instrumental variables and CACE (Complier Average Causal Effect) for noncompliance
  • Sensitivity analysis and bounds for missing data (e.g., Manski bounds)

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