← Back to Directory

Amazon interview notes, page 152 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·Technical Phone Screen

Jul 2026
Amazon BI Engineer interview with a SQL data quality question. Pretty straightforward anti-join scenario, nothing too wild, but it's the kind of thing that trips you up if you go blank on syntax under pressure.
  • Given an EMPLOYEE table and a DEPARTMENT table, write a SQL query to return all employees whose department ID does not exist in the DEPARTMENT table.

“Classic orphaned-record problem.”

View Post

Amazon·Software Engineer·Onsite - Coding / Algorithms

Jul 2026
Amazon new-grad SWE coding round with three problem areas: a from-scratch hash map, interval merging and intersection, and two water container variants. Pretty classic stuff but the breadth in one session was a lot to manage.
  • Implement a hash map from scratch supporting integer keys and values, with put, get, and remove operations, no built-in collections allowed. Also explain how you'd handle collisions and resize when the load factor gets too high.
  • Given an unsorted list of closed intervals, merge all overlapping ones and return a sorted list of non-overlapping intervals.
  • Follow-up to the interval merge: given two separate lists of pairwise-disjoint sorted intervals, find all intersections between intervals across the two lists.
  • Given an array of heights representing vertical lines, find two lines that form a container with the maximum possible water area.
  • Follow-up: given an elevation map as an array of heights, compute how much rainwater gets trapped in total.

“This took longer than I expected.”

View Post

Amazon·Machine Learning Engineer·Technical Phone Screen

Jul 2026
Amazon ML Engineer interview that went pretty deep into LLM fundamentals. Three big topic areas: Transformer internals, Mixture-of-Experts, and distributed training. More of a technical knowledge check than a coding round, which I wasn't fully expecting.
  • Walk me through the key components of the Transformer architecture and explain how self-attention works at a high level. Why does this architecture work better for language modeling than RNNs or LSTMs?
  • What problem does Mixture-of-Experts solve for large language models, and how does expert routing actually work? What are the main trade-offs you'd worry about?
  • Describe the different forms of parallelism used to train large models, explain what collective communication operations like all-reduce do, and give a concrete example of where all-reduce shows up when training a Transformer.

“I started with self-attention because I knew it well, explained the query-key-value thing, then talked about multi-head attention letting the model attend to multiple representation subspaces at once.”

View Post

Amazon·Software Engineer·Onsite - Multi Round

Jul 2026
Four rounds at Amazon London for a grad SDE role: two LP-plus-coding rounds, a GenAI fluency round where you literally solve a problem live with an AI tool on screen, and a Bar Raiser to close it out. Pretty intense but the GenAI round was genuinely unlike anything I'd done before.
  • Tell me about a time you realized midway through something that your approach or decision was wrong. What did you do?
  • Describe a goal that took a long time to achieve. What was your process, and what did you learn from it?
  • Implement a solution to a coding problem using an AI tool, with screen sharing on.
  • Coding problem focused on arrays and data structures, with follow-up questions on approach and complexity.
  • Design a solution using a doubly linked list combined with a hash map.
  • Bar Raiser behavioral round: extended discussion of past experiences and leadership principles.

“This one tripped me up a little because my first instinct was to pick a story where I looked good at the end.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026
Amazon SWE coding round with a class design problem. Pretty much one question the whole time, focused on OOP and randomness guarantees. Not the most intense interview but there were some gotchas in the implementation details.
  • Design a Card class that represents a standard 52-card deck. Implement a drawCard() method that returns a random card each call, with no repeats until all 52 cards have been drawn.

“My first instinct was to just use a random number and check if it was already drawn, which the interviewer let me finish before pointing out that gets slow as the deck empties.”

View Post

Amazon·Software Engineer·Online Assessment (OA)

Jul 2026
Sat through the Amazon OA and it was a mixed bag. The standard DSA problem went fine, all test cases passed. The AI coding section was rough though, only 2 of 6 TCs passed, which honestly has me wondering if I even get a rejection or just silence.
  • Solve a DSA problem as part of the online assessment.
  • Complete an AI-assisted coding challenge as part of the online assessment.

“This one went smoothly.”

View Post

Amazon·Machine Learning Engineer·Onsite - Behavioral / Leadership

Jul 2026
Behavioral round at Amazon for an ML Engineer role, basically one big question about decision-making under uncertainty. Pretty standard Amazon leadership stuff but the depth they expected was real.
  • Tell me about a time you made an important decision without having all the information you needed. Walk through the context, the options you weighed, the assumptions and risks you saw, how you gathered just enough data to move forward, and what happened. Looking back, what would you change and how did you handle the downside risk?

“This one is deceptively layered.”

View Post

Amazon·Machine Learning Engineer·Onsite - System Design / Architecture

Jul 2026
Amazon ML Engineer interview focused heavily on distributed training internals. Three questions, all technical, no fluff. The kind of round where you either know the mechanics or you don't.
  • Walk through the main parallelism strategies used when training models too large to fit on a single device. How does each one work, and what are the tradeoffs?
  • What are communication collectives like all-reduce, all-gather, reduce-scatter, and broadcast, and how do they actually show up in distributed training?
  • In tensor model parallelism, how do you split linear layers into column-parallel and row-parallel components? And what does it mean to alternate between column and row parallelism across layers, and why does that help?

“I started with data parallelism because it's the easiest to explain and bought myself time to think.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jul 2026
Amazon SWE interview with a tricky optimization problem about maximizing total bandwidth across ordered node pairs. The problem looked approachable at first but the constraints pushed you toward something smarter than brute force.
  • You have n processing nodes each with a bandwidth value. You need to select exactly streamCount distinct ordered pairs (i, j) of nodes, where each pair contributes bandwidth[i] + bandwidth[j] to a total sum. Pairs (i, j) and (j, i) count as different, and i == j is allowed. Maximize the total sum across all selected pairs. Your solution must handle n and streamCount up to around 10^5 efficiently, not by enumerating all n^2 pairs.

“The setup sounds like a greedy problem and it basically is, but I kept second-guessing myself on the i==j case and whether that changes the counting.”

View Post

Amazon·Data Scientist·Technical Phone Screen

Jul 2026
Amazon DS interview with a meaty causal inference case about a customer compensation pilot. The question had a lot of moving parts and felt more like a take-home problem compressed into a conversation.
  • Some regions piloted a program where customers affected by lost or damaged shipments receive a gift card instead of just a refund or replacement. How would you evaluate whether this program worked?
  • If you could redesign this as a randomized experiment from scratch, how would you set it up?

“This one sprawled fast.”

View Post