retail, devices, and AWS
Amazon Interview Questions (2026): 50 Real-World Questions + How to Answer
Amazon’s public leadership principles are the spine of the behavioral loop. Coding and design still matter, especially for SDE roles, but stories that do not map to a principle tend to fall flat. These prompts use the public principles as practice themes, not as a transcript. Public candidate writeups usually describe a recruiter screen, a coding or functional round, and a loop of behavioral interviews where each story is expected to land on a leadership principle. Write the principle in the margin of the story before you practice it. If the story does not need that principle, pick another story.
What will Amazon ask me?
Practice these prompts with an AI interview copilot, or download Luna for Chrome, Mac, and Windows.
Behavioral
Commonly reported
Tell me about a time you applied “Insist on the Highest Standards” when an Amazon.com deadline and a quality bar pulled in opposite directions.
How to approach. Open with the situation in one breath, then stay on your actions. At Amazon, name the tradeoff and tie the result to “Insist on the Highest Standards”.
Practice this with Luna
Behavioral
Commonly reported
Describe a teammate disagreement while working on something with the stakes of AWS. Where did “Have Backbone; Disagree and Commit” change your next step?
How to approach. Keep the story under two minutes. End with what changed for the user of a product like AWS, not with a moral.
Practice this with Luna
Behavioral
Commonly reported
Walk through a project you would happily retell in an Amazon behavioral round: the moment “Insist on the Highest Standards” cost you speed, and what you shipped anyway.
How to approach. Use a real miss if you have one. Amazon interviews reward the repair and the specific check you added afterward.
Practice this with Luna
Behavioral
Commonly reported
Give an example of pushing back on a plan because it conflicted with “Have Backbone; Disagree and Commit”. Use a situation as concrete as an Alexa launch.
How to approach. Say who disagreed with you and what you did with their objection. Do not flatten the story into “we aligned.”
Practice this with Luna
Behavioral
Commonly reported
Tell me about a miss. How did you repair it, and how would “Learn and Be Curious” show up in the way you explain that miss at Amazon?
How to approach. Put a number in the result — latency, revenue, tickets, or time — and admit what the number does not prove.
Practice this with Luna
Behavioral
Commonly reported
Describe influencing someone who did not report to you so an AWS-scale decision moved. Which part of the story shows “Earn Trust”?
How to approach. If you influenced without authority, name the ask, the resistance, and the artifact you left behind.
Practice this with Luna
Behavioral
Commonly reported
When did incomplete information force a call you could not postpone? Tie the story to “Bias for Action” without skipping what you checked first.
How to approach. State the assumption you tested before you acted. That is the part an Amazon interviewer can follow.
Practice this with Luna
Behavioral
Commonly reported
Tell me about raising the quality of someone else’s work. How would you connect that story to “Hire and Develop the Best” for an Amazon interviewer?
How to approach. Credit the teammate whose work you improved, then describe the standard you held and how you showed it.
Practice this with Luna
Behavioral
Commonly reported
Describe a time customer impact — the kind Amazon.com users would feel — made you change scope. What did “Customer Obsession” require you to drop?
How to approach. Start from the user of Amazon.com, then the constraint, then your call. Skip the company history.
Practice this with Luna
Behavioral
Commonly reported
Tell me about saying no to a shortcut that would have looked good in a status update. How does that choice reflect “Insist on the Highest Standards”?
How to approach. Practice the “no” out loud. Include the shortcut you refused and the slower path you chose instead.
Practice this with Luna
Coding / DSA
Commonly reported
Amazon coding theme to practice: given a stream of inventory reservation ids, return the most frequent few without sorting the whole history. State time and space, then name one edge case.
How to approach. Repeat the input and output in your own words, then give the naive cost. Only then reach for a better structure for inventory reservation ids.
Practice this with Luna
Coding / DSA
Commonly reported
Practice prompt in an Amazon coding round: detect whether a sequence of package scan events repeats in a cycle, and describe the smallest window that proves it. Explain the invariant before you code.
How to approach. Write the invariant on the side before the loop. Amazon rounds go better when the interviewer can see what stays true.
Practice this with Luna
Coding / DSA
Commonly reported
Commonly discussed Amazon data-structure theme: merge overlapping ranges drawn from Prime video sessions and report the covered span. Talk through a correct pass, then a tighter one.
How to approach. Dry-run three cases: empty, one element, and the boundary you are afraid of. Talk while you do it.
Practice this with Luna
Coding / DSA
Commonly reported
Representative Amazon graph prompt: find a path across Alexa utterance ids when some links are closed. Say which search you would use and why.
How to approach. Name the search or traversal and the reason. If you switch approaches, say what the first one failed to handle.
Practice this with Luna
Coding / DSA
Commonly reported
Amazon coding practice: design an iterator over AWS API request ids that stays correct while new items arrive. Cover an empty start and a burst in the middle.
How to approach. If new data arrives while you iterate, say whether you snapshot or follow updates. Pick one and stick to it.
Practice this with Luna
Coding / DSA
Commonly reported
Practice this Amazon style prompt: keep a running median over the latest window of review helpfulness votes. Explain the two structures you would hold and their costs.
How to approach. For a running aggregate, say the window, the update cost, and what you evict. A diagram of two structures is enough.
Practice this with Luna
Coding / DSA
Commonly reported
Coding theme associated with Amazon interviews: group fulfillment center tasks that share a signature. Describe the hash you would explain out loud and what collision means here.
How to approach. Explain the key you hash and what a collision means for this data. Do not jump to library names.
Practice this with Luna
Coding / DSA
Commonly reported
Amazon scheduling-style prompt: assign subscribe-and-save renewals to a fixed pool of workers so the finish time is as small as you can justify. Start with a greedy idea and say where it fails.
How to approach. Start greedy, then show one input where greedy is wrong if it is wrong. That correction is the answer.
Practice this with Luna
Coding / DSA
Commonly reported
Representative Amazon search prompt: locate an id inside a rotated, mostly sorted list of returns reason codes. Narrate the boundary checks before you write the loop.
How to approach. Narrate the low and high bounds before the first iteration. Off-by-one is the usual miss on rotated data.
Practice this with Luna
Coding / DSA
Commonly reported
Amazon coding theme: admit a burst of CloudWatch alarm bursts only while each key stays under a fixed per-minute budget. Define the key, the window, and how you would test a boundary.
How to approach. Define the identity you rate-limit, the clock, and the response when the budget is spent. Test the exact boundary.
Practice this with Luna
System design
Commonly reported
System design practice for Amazon: sketch a checkout inventory hold for read-heavy traffic. Start with the API and the data you store, then add one cache and say what it does not cover.
How to approach. Users and API first. For a checkout inventory hold, say the one request you must make fast and the one you can delay.
Practice this with Luna
System design
Commonly reported
Design package tracking for a region the way an Amazon round expects: a single-region failure should not take the product down. Say what you would refuse to build in the first conversation.
How to approach. Draw one failure domain and what still works. Amazon interviewers listen for what you explicitly will not build yet.
Practice this with Luna
System design
Commonly reported
Capacity exercise themed on Amazon: estimate a Prime video start-up path before you pick a database. Use rough numbers and label every assumption.
How to approach. Put the capacity math on the board with units. Round numbers are fine if every assumption is labeled.
Practice this with Luna
System design
Commonly reported
Design prompt for Amazon: Alexa skill routing needs an API that a client can retry safely. Define the idempotency key and what the user sees on a timeout.
How to approach. Retries need an idempotency story. Say what the client sends the second time and what the user sees.
Practice this with Luna
System design
Commonly reported
Practice designing an S3-style object index for Amazon. Separate the hot path from the reporting path, and name the queue or log that joins them.
How to approach. Split the hot path from analytics. Name the log or queue in one sentence and move on.
Practice this with Luna
System design
Commonly reported
Amazon design theme: review ranking on a product page must stay fast for the common case and honest when data is stale. Where would you show staleness to the user?
How to approach. If a cache can be stale, say who notices and how the product tells the truth. Do not hide staleness.
Practice this with Luna
System design
Commonly reported
Sketch a returns label service and focus on authorization. Who at Amazon-scale is allowed to read, write, and audit?
How to approach. List readers, writers, and auditors before you pick a database. Authorization is part of the design, not a footer.
Practice this with Luna
System design
Commonly reported
Design subscribe-and-save renewal with a migration story. How would you move from the current store to the next one without a long freeze?
How to approach. Describe the migration in two phases and the rollback. A long freeze is a smell; say so.
Practice this with Luna
System design
Commonly reported
Practice prompt: a multi-region product catalog has a sharp traffic spike. What do you shed first, and which Amazon user promise do you refuse to shed?
How to approach. Name the load you shed first and the promise you keep. Tie that promise to Amazon.com.
Practice this with Luna
System design
Commonly reported
Design on-call paging for a tier-1 AWS API for multi-tenant fairness. How do you stop one noisy customer from crowding out everyone else?
How to approach. Noisy-neighbor control is a product decision. Say the fairness unit — tenant, user, or key — and the limit.
Practice this with Luna
Role-specific
Commonly reported
Amazon software development engineer prompt: walk through a decision you owned on a product like Amazon.com. What did you measure after it shipped?
How to approach. Answer as a software development engineer who owned an outcome. Include the metric you checked after an Amazon.com change shipped.
Practice this with Luna
Role-specific
Commonly reported
For a frontend engineer loop at Amazon, how would you prioritize two AWS requests that both look urgent to their sponsors?
How to approach. Prioritization needs a rule, not a vibe. Say whose request waited and how you told them.
Practice this with Luna
Role-specific
Commonly reported
Role practice: as a product manager at Amazon, tell me about a metric you stopped reporting because it was making the team worse.
How to approach. If a metric made the team worse, say what you replaced it with. Amazon hears the replacement, not the complaint.
Practice this with Luna
Role-specific
Commonly reported
In an Amazon data scientist interview, explain a cross-team dependency around Alexa and how you unblocked it without taking over their work.
How to approach. Describe the dependency, the ask, and what you refused to take over. Cross-team stories die when you become the hero.
Practice this with Luna
Role-specific
Commonly reported
engineering manager prompt themed on Amazon: describe the smallest version of an Amazon.com change you would ship to learn, and what would make you stop.
How to approach. Define the smallest shippable slice and the stop condition. Learning is the point of the first version.
Practice this with Luna
Role-specific
Commonly reported
How would you, as a solutions architect, explain an Amazon-style tradeoff between reliability and an AWS feature the business wants this quarter?
How to approach. Reliability versus an AWS feature is a tradeoff with a user attached. Name the user of AWS and the failure you will not accept.
Practice this with Luna
Role-specific
Commonly reported
Amazon technical program manager question to practice: tell me about feedback you acted on that changed the design, not just the wording.
How to approach. Quote the feedback in one line, then the design change. Skip the feelings recap.
Practice this with Luna
Role-specific
Commonly reported
As a support engineer interviewing at Amazon, how do you decide what not to build when Alexa already has a long backlog?
How to approach. A backlog story should end with something you cut. Say the criterion you used.
Practice this with Luna
Role-specific
Commonly reported
Role-specific prompt: describe how you would onboard onto Amazon.com at Amazon in your first thirty days and what you would not pretend to know.
How to approach. A thirty-day plan should include what you will not pretend to know. Name the person you would shadow.
Practice this with Luna
Role-specific
Commonly reported
For an Amazon machine learning engineer round, walk through a postmortem you led. What changed in the system, and what changed in the team?
How to approach. A postmortem needs a system change and a team change. One of them is not a slide.
Practice this with Luna
Culture & values
Commonly reported
How would you explain “Customer Obsession” to a new teammate using a decision that affected people who rely on something like AWS?
How to approach. Translate “Customer Obsession” into a decision with a cost. A definition without a cost sounds memorized.
Practice this with Luna
Culture & values
Commonly reported
Culture prompt for Amazon: where does “Insist on the Highest Standards” conflict with moving fast, and how have you handled that tension?
How to approach. Name the tension with speed honestly. Amazon culture questions are about the tradeoff, not the slogan.
Practice this with Luna
Culture & values
Commonly reported
Which of Amazon's public values is hardest for you — start with “Insist on the Highest Standards” — and what do you do when you notice yourself skipping it?
How to approach. Pick the value you actually find hard. Interviewers can hear a humblebrag.
Practice this with Luna
Culture & values
Commonly reported
Describe a hiring or collaboration choice that would show an interviewer you take “Hire and Develop the Best” seriously, not as a poster.
How to approach. Use a collaboration or hiring moment, not a speech about culture. One scene is enough.
Practice this with Luna
Culture & values
Commonly reported
How should an AWS team at a company like Amazon behave when “Customer Obsession” and a revenue target disagree?
How to approach. If revenue and the value disagree, say who you would talk to and what evidence you would bring about Amazon.com.
Practice this with Luna
Culture & values
Commonly reported
Culture practice: tell a short story that would convince an Amazon interviewer you can disagree and still protect “Have Backbone; Disagree and Commit”.
How to approach. Disagreement needs a sentence you can say in the room. Practice that sentence, then the relationship you kept.
Practice this with Luna
Culture & values
Commonly reported
What would you want a skip-level to notice about how you live “Ownership” on an Alexa project?
How to approach. Skip-level answers should be observable behavior, not intent. What would they have seen?
Practice this with Luna
Culture & values
Commonly reported
If a process at Amazon made “Invent and Simplify” harder, what would you change first, and what evidence would you bring?
How to approach. Process complaints need a first change and a way to know it worked. Keep it small.
Practice this with Luna
Culture & values
Commonly reported
How do you keep “Customer Obsession” in the room when the conversation is only about the AWS roadmap?
How to approach. Put the value in the roadmap conversation as a constraint, not a preamble. Then make the cut.
Practice this with Luna
Culture & values
Commonly reported
Culture prompt: name a behavior you would not tolerate on an Amazon team because it breaks “Insist on the Highest Standards”, and how you would say that out loud.
How to approach. “What you would not tolerate” needs a calm sentence and a next step. No speeches about toxicity.
Practice this with Luna