← Back to Directory

Amazon interview notes, page 18 of 159

FAANG / Big Tech

Most of these Amazon write-ups come from Software Engineers (1,092), then Data Scientists (182) and Product Managers (165). The most reported rounds are the technical phone screen (618), the behavioral / leadership round (309), the online assessment (228) and the system design round (159).

1581 interview notes · updated Oct 2026

Amazon·Software Engineer·Online Assessment (OA)

Apr 2026
Amazon SWE online assessment with a greedy/backtracking sequence construction problem. Pretty niche problem type, not your typical LC medium.
  • Given an integer array of values and a binary string representing which indices are currently available, build a sequence of length m by repeatedly picking an available index and appending its value. After each pick, any index immediately to the right of an available index gets unlocked. Return the lexicographically largest sequence possible.

“This one took me a while to even parse the problem statement.”

View Post

Amazon·Software Engineer·Onsite - System Design / Architecture

Apr 2026
Amazon system design round for a software engineer role. The whole session was basically one big design question about tracking perishable inventory across a retail network, with the interviewer pushing hard on scaling and data store justification. Felt like I was playing catch-up the whole time.
  • Design a backend service that tracks perishable items across a retail store network and can report where any given item is at any point in time, including whether it's on a shelf or has been moved to a storage area after expiration.
  • What clarifying questions would you ask before diving into the design? Specifically around what an 'item' means, where expiration events come from, and how fresh the read results need to be.
  • How would you design the GET endpoints for this system, and what does the underlying data model look like to support both a locate-by-item lookup and a per-store listing filtered by status?
  • Should you use a relational database or a NoSQL store for this system, and why? If you say consistency is the reason to go relational, justify it on more than just ACID.
  • With millions of items, running a full table scan every few minutes to find what has expired will destroy your database. How do you design the expiration pipeline to avoid that?
  • How does your read path scale to handle millions of requests, potentially in spikes?
  • How would you handle a move event that arrives out of order or gets submitted twice?
  • If the product team wants a full location history per item as an audit trail, how does your data model change and what's the storage cost implication?
  • How do you keep the locate-by-ID cache consistent when a move event is recorded? Walk through the trade-offs between write-through, invalidate-on-write, and a short TTL approach.

“I started with the API shape which was probably the right call, but I spent too long debating the data model before nailing down what the actual read patterns were.”

View Post

Amazon·Machine Learning Engineer·Technical Phone Screen

Apr 2026
Mixed round at Amazon for an MLE role, covering both applied ML system design and theory. The search relevance scenario was meaty but the theory follow-ups came out of nowhere and I wasn't fully prepared for the second one.
  • How would you build a relevance prediction system for Amazon search results? Walk through your feature engineering and modeling choices.
  • Can you explain the bias-variance trade-off?
  • What is the double-descent phenomenon, and why does test error sometimes drop again after you add more parameters or train past a certain threshold?

“This is the kind of question where you can go a lot of directions and that's kind of the problem.”

View Post

Amazon·Data Scientist·Technical Phone Screen

Apr 2026
Amazon Data Scientist interview with a technical coding round focused on number theory and algorithmic efficiency. The questions escalated pretty quickly from a basic sieve implementation into memory-constrained variants and then a curveball about arbitrary input arrays. Solid prep on classical algorithms is a must here.
  • Write a function that returns all prime numbers from 1 to n (inclusive), where n can be as large as 10^7. The result should be in ascending order and the solution should be efficient in both time and memory.
  • Now suppose n can be up to 10^12 and memory is a real constraint. How would you adapt your approach? Walk through the complexity.
  • Instead of the full range 1 to n, you're given an unsorted array of distinct integers drawn from 1 to n. How would you find which of those values are prime?

“I went straight for the Sieve of Eratosthenes and it felt fine.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Apr 2026
Got a data visualization question during what felt like a product analytics screen for a BA role at Amazon. Just the one question but it had enough layers to it that I fumbled around for longer than I'd like to admit.
  • You work at a rideshare company. How would you visually represent the correlation between ride duration, fare, and ratings?

“Three variables is where I got tripped up.”

View Post

Amazon·Data Scientist·Technical Phone Screen

Apr 2026
SQL-heavy technical screen for a Data Scientist role at Amazon. Three questions, all focused on querying event and session tables, nothing behavioral at all. Pretty straightforward if you're comfortable with GROUP BY and window functions, but the last one tripped me up a bit.
  • Given a table of user events, write a SQL query to count the total number of events per user.
  • Using the same events table, return each user's earliest and latest event timestamps.
  • Join the events and sessions tables to find the user whose single session has the longest duration, and return both the user ID and that duration.

“Easiest of the three.”

View Post

Amazon·Product Manager·Onsite - Behavioral / Leadership

Apr 2026
Amazon PM interview with a metrics-focused behavioral question. Pretty standard loop but this one made me think harder than I expected.
  • Walk me through a decision you made that was driven by a metric you were actively tracking.

“I fumbled the setup a bit.”

View Post

Amazon·Data Scientist·Onsite - System Design / Architecture

Apr 2026
One question, no small talk. Amazon asked me to design a full operations dashboard for a delivery-station manager and just let me run with it for the whole session. Dense, exhausting, and weirdly fun once I stopped panicking.
  • Design an operations dashboard for a delivery-station manager to monitor same-day delivery performance. Cover which metrics to show, which visualizations to use and why, the layout, interaction design, alert thresholds, how you'd validate usefulness, and how you'd handle 50k+ events per day without sacrificing responsiveness or tail accuracy.

“This was basically a 45-minute design session disguised as one question.”

View Post

Amazon·Software Engineer·Onsite - Behavioral / Leadership

Apr 2026
Behavioral round at Amazon for a SWE role. Pretty short on details but the question about handling critical feedback is a classic Amazon ask, and it's one I clearly hadn't prepped enough for.
  • How open are you to critical feedback, and can you walk me through a specific time you received it?

“I fumbled this more than I expected.”

View Post

Amazon·Software Engineer·Onsite - Multi Round

Apr 2026
Went through a four-round loop for an Amazon new grad SDE role and mostly held it together until the very last round, where exhaustion got the better of me and I coded the wrong approach despite knowing the right one. Now sitting here wondering if one messy screen tanks an otherwise strong loop.
  • Given a coding problem, walk through multiple approaches and implement the one you think is best.
  • Design the architecture and code flow for a given system or feature.

“This is where I fumbled badly.”

View Post