Amazon Interview Questions — Page 151

Amazon·Software Engineer·Onsite - System Design / Architecture

Jul 2026

Amazon system design round focused entirely on a document and tag management system, starting from basic class design and scaling up to full concurrency. Pretty intense for a single question that kept growing.

Design a system to manage documents and tags, including class interfaces, core data structures, and support for operations like creating/deleting documents and tags, adding/removing tags on documents, listing documents by tag, listing tags by document, and querying documents by the intersection of multiple tags.
How would you extend the document and tag management design to be thread-safe under concurrent reads and writes? Cover locking granularity, consistency guarantees, deadlock avoidance, and how you'd test for concurrency correctness.

“I started with two hashmaps, one mapping document IDs to their tag sets and another mapping tags to document sets.”

View Post

Amazon·Software Engineer·Onsite - Coding / Algorithms

Jul 2026

Amazon SWE coding round with a string compression problem that looked deceptively simple but had enough edge cases to keep you honest. The encoding rules were a bit unusual and I had to think carefully about when to apply RLE versus just leaving characters as-is.

Given a string, implement an encoder and decoder using a hybrid run-length encoding strategy: apply RLE only when there are 8 or more consecutive identical characters, otherwise keep the raw characters. Any leftover run that doesn't hit 8 should still use RLE if it can.

“The tricky part was the 'bbbbbbbbbbb' case, 11 b's that should come out as '8b3b'.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026

Amazon SQL interview, one meaty question about aggregating click data across a product hierarchy. The whole thing revolved around a single problem but they kept pulling the thread until I was talking about indexing and multi-department rollups.

Given a Products table with a department/category/subcategory hierarchy and a ClickLog table tracking user clicks on products, write SQL to count distinct users who clicked any product in a specified department within a given time range. Then explain how you'd handle indexing and partitioning at scale, and extend the query to return results across all departments.

“The core join was fine, just joining ClickLog to Products on product_id and filtering by department and a timestamp range, then COUNT(DISTINCT user_id).”

View Post

Amazon·Product Manager·Onsite - Product Sense / Strategy

Jul 2026

Amazon PM interview with a product launch question that gave you three different service types to pick from. Pretty open-ended, which sounds nice until you realize you have to structure six different dimensions for each one.

Pick one of these three potential Amazon services (a table-booking app, a flower-delivery service, or an advertising platform) and walk through how you'd launch it. Cover target users, pain points, value proposition, MVP features, success metrics, and go-to-market strategy.

“I went with the advertising platform because I figured I had the most to say about it, but in hindsight the table-booking app might have been the sharper choice since it has more obvious Amazon ecosystem angles.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026

Amazon SWE coding round with one question about friend-based product recommendations. Pretty straightforward graph/set traversal problem but the data structure setup took me a minute to parse.

Given a user object that contains a list of friends (each also a user object with purchases) and a list of the user's own purchases, write a function that returns the names of products bought by at least one friend but not by the user.

“The core logic isn't hard once you see it: collect all product IDs from friends' purchases into a set, subtract the user's own purchases, then map the remaining IDs to product names.”

View Post

Amazon·Software Engineer·Online Assessment (OA)

Jul 2026

Amazon OA for a SWE role, one coding problem about string scoring with an optional character swap. Pretty algorithmic, felt like a mid-level screen.

Given a string, compute a score where each character is worth 1 point and each substring where adjacent characters differ by at most 1 (alphabetically) earns a bonus point. You can change at most one character to maximize the total score. What's the highest score achievable?

“The base scoring wasn't hard to wrap my head around but the 'change one character' part is where I fumbled for a bit.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026

Amazon SDE Intern technical phone screen that ran about 50 minutes, covering two projects, a couple LP questions, and two leetcode-style problems. The candidate got through both coding problems but only wrote code for the first one, which left them second-guessing themselves afterward.

Walk me through one of your projects.
Tell me about a time you worked under a tight deadline.
Have you worked in a team? What was your role?
Describe a time you had a conflict with a teammate over an idea or approach.
Given an array of bloom days for flowers, find the minimum number of days needed to make m bouquets of k adjacent flowers. (LeetCode: Minimum Number of Days to Make m Bouquets)
What is a BST, a full binary tree, and a complete binary tree?
Two nodes in a BST have been swapped. Recover the original BST. (LeetCode: Recover Binary Search Tree)

“He asked about one project, I explained it, then he circled back to the same project again after I described a second one.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026

Amazon SWE interview with a circular route shortest path problem. Pretty standard algorithmic stuff but the prefix sum optimization is what they're really looking for, not just the brute force.

You have a circular route with n stops and an array of clockwise distances between consecutive stops. Given a list of queries each asking for the shortest distance between two stops (you can go either direction), return the sum of shortest distances across all queries. Design an efficient solution for large inputs.

“My first instinct was to just compute the clockwise sum for each query by looping through the relevant segment, then subtract from total to get the counterclockwise distance.”

View Post

Amazon·Software Engineer·Online Assessment (OA)

Jul 2026

Went through Amazon's Work Simulation OA for a software engineer role. It's a mix of behavioral ratings and system design scenarios, which I wasn't fully expecting. Some of the technical questions were genuinely interesting, others felt like they were testing whether you'd read the docs.

What is the purpose of the Work Simulation test and how is it structured across its five modules?
In the Work Simulation, you're given general workplace scenarios and a list of possible actions. Rate how effective each action would be.
For a real-time voting service built on Amazon Voice, which vote-storage strategy is most effective from the options provided?
You're designing a new SaaS inventory management system. Based on a set of product emails, which design actions should you prioritize next?
Compare different approaches for storing image thumbnails in the inventory management system and rate their effectiveness.
For a traffic-video service that uses queues, prioritize the design actions needed to define a message format covering versioning, binary serialization, and checksums.
What approaches would you recommend for reliably transmitting very large camera messages from a traffic-video service to a central service over an unreliable network?
What measures would you put in place to monitor message loss and improve resilience in the traffic-video service?
What strategies would you use to ensure high availability for an inventory management system being launched globally?

“I went in blind on this and had to figure out the structure as I went.”

View Post

Amazon·Software Engineer·Onsite - System Design / Architecture

Jul 2026

Amazon system design round focused almost entirely on distributed key-value store replication. No fluff, no warmup, just straight into quorums and failure modes for the whole session.

Design a distributed key-value store that replicates data across multiple replicas. Cover the APIs, consistency goals, partitioning strategy, quorum rules, and how you handle node failures, network partitions, and write conflicts.

“I started with Get and Put APIs which was fine, but then I spent way too long on partitioning before they nudged me toward quorums.”

View Post