Amazon Site Reliability Engineer Interview Questions (2026)

1. 1Online Assessment (SDE OA)60 minMedium

   Two timed coding problems plus a workplace-simulation and logic section; the main gate for freshers and India volume hiring.

2. 2Phone screen45 minMedium

   One coding problem plus 1-2 Leadership Principle STAR questions with an SDE.

3. 3Coding loop round60 minMedium

   DSA problem to working code, followed by assigned-LP behavioral questions in STAR format.

4. 4System design loop round60 minHard

   Design an Amazon-scale service with capacity math, plus LPs; low-level/OOD design substitutes for junior candidates.

5. 5Hiring Manager round45 minMedium

   Team fit, project deep dives, and Deliver Results/Bias for Action stories with the manager you would report to.

6. 6Bar Raiser60 minHard

   An interviewer from outside the team stress-tests LP stories and overall bar with the hardest cross-examination of the loop; holds veto.

1. Q1

   Explain how to apply IAM least privilege and roles in AWS for this production context: high-volume customer-facing retail/service platform. What problem does it solve, and where can it fail? Assume the target company is Amazon and the priority is rapid incident detection and mitigation

   EasyTechnical ScreenAWS

   How to answer:

   A strong answer defines the mechanism, names the operational boundary, and states when it is the right tool. For IAM least privilege and roles: Use IAM roles over long-lived keys, grant least privilege with scoped actions/resources, add permission boundaries when needed, and audit with CloudTrail. Validate with access analyzer or policy simulation before rollout. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Mention how you would validate the behavior in a non-production environment and what metric proves it is working.

2. Q2

   How would you make CloudWatch, CloudTrail, and EventBridge resilient while keeping cost and operational complexity under control? Assume the target company is Amazon and the priority is minimal operational toil for a small platform team

   MediumAWS + Terraform Deep DiveAWS

   How to answer:

   A strong answer maps the design to SLO, RTO/RPO, blast radius, and recovery tests. For CloudWatch, CloudTrail, and EventBridge: Use CloudWatch for metrics/logs/alarms, CloudTrail for audit trails, and EventBridge for event-driven automation. Tie alarms to actionable runbooks and avoid alerts with no owner or response. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Prefer simple resilient patterns first: health checks, retries with backoff, graceful degradation, backups, redundancy, and regular game days.

3. Q3

   You need to migrate legacy production usage of Cost optimization and quotas in this context: high-volume customer-facing retail/service platform, without downtime. How would you plan and execute it? Assume the target company is Amazon and the priority is rapid incident detection and mitigation. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

   HardAWS + Terraform Deep DiveAWS

   How to answer:

   A strong answer uses inventory, compatibility, staged rollout, verification, and rollback. For Cost optimization and quotas: Tag resources, right-size compute, use savings/reservations for stable workloads, set budgets, and monitor service quotas. Reliability should be evaluated against explicit business impact, not unlimited spend. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Run dual-write or shadow traffic where appropriate, compare outputs, migrate cohorts, monitor error budgets, and keep a rollback window.

4. Q4

   Compare two viable approaches to IAM least privilege and roles for this context: high-volume customer-facing retail/service platform. What would make you choose one over the other? Assume the target company is Amazon and the priority is low-latency global user experience. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

   MediumAWS + Terraform Deep DiveAWS

   How to answer:

   A strong answer compares constraints instead of declaring a universal best practice. For IAM least privilege and roles: Use IAM roles over long-lived keys, grant least privilege with scoped actions/resources, add permission boundaries when needed, and audit with CloudTrail. Validate with access analyzer or policy simulation before rollout. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. State the decision criteria: risk, team expertise, operational load, lock-in, cost, compliance, and reversibility.

5. Q5

   Outline the steps to implement CloudWatch, CloudTrail, and EventBridge safely for this production context: high-volume customer-facing retail/service platform. Include validation, rollout, and rollback. Assume the target company is Amazon and the priority is high deployment velocity without increasing incidents

   MediumAWS + Terraform Deep DiveAWS

   How to answer:

   A strong answer breaks work into small reversible changes with automated checks. For CloudWatch, CloudTrail, and EventBridge: Use CloudWatch for metrics/logs/alarms, CloudTrail for audit trails, and EventBridge for event-driven automation. Tie alarms to actionable runbooks and avoid alerts with no owner or response. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Use peer-reviewed code, tests, policy checks, staged rollout, observability, and a rollback plan before widening scope.

6. Q6

   Design a production AWS approach using VPC subnet design and routing. The service context is high-volume customer-facing retail/service platform, and it must handle cost control during unpredictable traffic spikes. How do you structure the solution and tradeoffs? Assume the target company is Amazon and the priority is cost control during unpredictable traffic spikes. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

   MediumAWS + Terraform Deep DiveAWS

   How to answer:

   A strong answer turns requirements into architecture, controls, automation, and measurable failure handling. For VPC subnet design and routing: Separate public and private subnets across Availability Zones, keep route tables explicit, use NAT only where needed, and prove connectivity with flow logs and route analysis. Design for blast-radius containment. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Include IaC, CI/CD, monitoring, security boundaries, capacity assumptions, and the exact rollback or failover path.

7. Q7

   Explain how to apply S3 durability, access, and lifecycle in AWS for this production context: high-volume customer-facing retail/service platform. What problem does it solve, and where can it fail? Assume the target company is Amazon and the priority is 99.9% availability with fast rollback

   EasyTechnical ScreenAWS

   How to answer:

   A strong answer defines the mechanism, names the operational boundary, and states when it is the right tool. For S3 durability, access, and lifecycle: Use bucket policies, block public access, KMS encryption where required, versioning, lifecycle rules, and replication only for defined RPO/RTO or compliance needs. Monitor access logs and object-level events when risk warrants it. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Mention how you would validate the behavior in a non-production environment and what metric proves it is working.

8. Q8

   What security risks commonly appear around pipeline secrets, and how would you reduce them in production? Assume the target company is Amazon and the priority is tenant isolation for enterprise customers

   MediumCI/CD and Deployment AutomationCI/CD

   How to answer:

   A strong answer assumes misconfiguration will happen and designs guardrails plus detection. For pipeline secrets: Store secrets in platform secret stores, scope them by environment, and prefer short-lived federated credentials. Never print secrets in logs. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Apply least privilege, encryption, secret handling, audit logs, vulnerability management, and automated policy enforcement.

9. Q9

   Tell me about a time you used CI/CD or build caching to improve customer obsession, ownership, frugality, and operational excellence. What did you measure and learn? Assume the target company is Amazon and the priority is low-latency global user experience. Use a Amazon-style example and include measurable production impact

   EasyLeadership Principles / Bar RaiserCI/CD

   How to answer:

   A strong answer uses STAR: situation, task, action, result, and lesson learned. For build caching: Cache dependencies with precise keys and invalidation rules. Balance speed with reproducibility and avoid caching sensitive files. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Quantify impact with latency, availability, cost, deployment frequency, MTTR, defect rate, or toil reduction.

10. Q10

    Design a production CI/CD approach using blue/green deployments. The service context is high-volume customer-facing retail/service platform, and it must handle tenant isolation for enterprise customers. How do you structure the solution and tradeoffs? Assume the target company is Amazon and the priority is tenant isolation for enterprise customers. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

    MediumCI/CD and Deployment AutomationCI/CD

    How to answer:

    A strong answer turns requirements into architecture, controls, automation, and measurable failure handling. For blue/green deployments: Deploy the new version beside the old one, validate it, switch traffic, and keep rollback simple. Watch for data/schema compatibility before switching. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Include IaC, CI/CD, monitoring, security boundaries, capacity assumptions, and the exact rollback or failover path.

11. Q11

    During on-call, user impact points toward canary deployments. What do you do in the first 15 minutes, the next hour, and after recovery? Assume the target company is Amazon and the priority is rapid incident detection and mitigation

    MediumCI/CD and Deployment AutomationCI/CD

    How to answer:

    A strong answer prioritizes mitigation, communication, evidence, and prevention. For canary deployments: Shift a small percentage of traffic, compare golden signals, and automate promotion or rollback. Use canaries only when telemetry can detect regressions quickly. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Declare severity, assign roles, mitigate first, communicate cadence, preserve timeline, and convert root cause into tested corrective actions.

12. Q12

    You need to migrate legacy production usage of test reliability and flaky tests in this context: high-volume customer-facing retail/service platform, without downtime. How would you plan and execute it? Assume the target company is Amazon and the priority is secure delivery of regulated workloads. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

    MediumCI/CD and Deployment AutomationCI/CD

    How to answer:

    A strong answer uses inventory, compatibility, staged rollout, verification, and rollback. For test reliability and flaky tests: Quarantine or fix flaky tests, track flake rate, and avoid normalizing reruns. Flaky pipelines erode trust in deployment safety. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Run dual-write or shadow traffic where appropriate, compare outputs, migrate cohorts, monitor error budgets, and keep a rollback window.

13. Q13

    Workload in this production context (high-volume customer-facing retail/service platform) grows 10x. How would you scale and protect artifact versioning? Assume the target company is Amazon and the priority is tenant isolation for enterprise customers

    MediumCI/CD and Deployment AutomationCI/CD

    How to answer:

    A strong answer measures the bottleneck before adding capacity and protects downstream dependencies. For artifact versioning: Build once and promote the same immutable artifact through environments. Attach metadata such as commit SHA, SBOM, test results, and provenance. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Use load tests, autoscaling policies, queue/backpressure controls, quota reviews, and cost alarms; confirm the user-facing SLI improves.

14. Q14

    Compare two viable approaches to branching and release strategy for this context: high-volume customer-facing retail/service platform. What would make you choose one over the other? Assume the target company is Amazon and the priority is rapid incident detection and mitigation. Frame the answer for an interview loop where expect data-backed tradeoffs, failure analysis, and STAR examples tied to Leadership Principles

    MediumCI/CD and Deployment AutomationCI/CD

    How to answer:

    A strong answer compares constraints instead of declaring a universal best practice. For branching and release strategy: Keep branches short-lived when possible, use protected branches, and define release channels clearly. Match strategy to deployment frequency and regulatory requirements. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. State the decision criteria: risk, team expertise, operational load, lock-in, cost, compliance, and reversibility.

15. Q15

    A production service (high-volume customer-facing retail/service platform) is failing after a change involving layer caching and build context. Walk through how you would investigate, mitigate, and fix it. Assume the target company is Amazon and the priority is low-latency global user experience

    MediumDocker + Kubernetes Platform RoundDocker

    How to answer:

    A strong answer starts with impact, recent changes, and evidence before changing production. For layer caching and build context: Order Dockerfile steps from least to most frequently changing, use .dockerignore, and copy dependency manifests before source when possible. Keep cache useful but avoid stale vulnerable bases. In a high-volume customer-facing retail/service platform, tie the decision to customer obsession, ownership, frugality, and operational excellence. Check logs, metrics, events, deployment diffs, permissions, dependencies, and rollback options; then write a durable fix and postmortem item.

Practice these with instant AI feedback in a live mock interview → Start a Amazon Site Reliability Engineer mock

Prepare 8-12 STAR stories mapped to Leadership Principles; expect a Bar Raiser; quantify impact

Indicative Site Reliability Engineer pay in India: ~₹12–52 LPA (role-level range, not a Amazon-specific figure).

Other roles at Amazon

Site Reliability Engineer interviews at other companies