← Back to Directory

Amazon interview notes, page 135 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

Jun 2026
Amazon SWE coding round, one question on tree traversal. Pretty short session, nothing wild.
  • Implement an inorder traversal for an N-ary tree.

“Tricky part is that inorder for N-ary isn't as cleanly defined as it is for binary trees.”

View Post

Amazon·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Amazon ML engineer interview that was basically one big coding question: implement a GPT-style decoder-only transformer from scratch, modular, four classes, correct shapes throughout. No behavioral stuff from what I remember, just pure implementation.
  • Implement a multi-head self-attention module with a causal mask. The class should take a tensor of shape (B, T, d_model), project into Q/K/V, split into heads, compute scaled dot-product attention with a causal mask, and return the concatenated output projected back to (B, T, d_model).
  • Implement a position-wise feed-forward network class that applies two linear layers with a nonlinear activation in between, independently at each sequence position.
  • Implement a single decoder block that combines masked self-attention and the feed-forward network using pre-norm residual connections (LayerNorm applied before each sub-layer).
  • Implement the full GPT model class: token embeddings plus positional encodings, a stack of N decoder layers, and a final linear projection to vocabulary logits. The forward pass should take integer token IDs and return logits of shape (B, T, V).

“The causal mask part is where I fumbled a bit.”

View Post

Amazon·Data Scientist·Onsite - System Design / Architecture

Jun 2026
Amazon data scientist interview, system design round focused entirely on building a spam detection pipeline from scratch. One long question, no breaks, they just kept pushing on every layer until you either had an answer or you didn't.
  • Design a full end-to-end email spam detection system with real-time scoring under 50ms p99 latency, high precision on hard blocks, and the ability to adapt as adversaries evolve their tactics.

“This was basically seven questions stitched into one.”

View Post

Amazon·Product Manager·Onsite - Behavioral / Leadership

Jun 2026
PM loop at Amazon, behavioral focus. Just one question shared but it's a classic one that trips people up more than you'd expect.
  • Describe a time you had to make a major decision without your manager's input or approval.

“I went straight into the story without setting up the stakes first and I think that hurt me.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jun 2026
Amazon SWE interview with a grid pathfinding problem. Pretty standard Dijkstra territory but they pushed on implementation details and path reconstruction which is where things got interesting.
  • Given an m×n grid with nonnegative cell costs and some cells blocked, find the minimum cost path from a start cell to a target cell moving in four directions, or return -1 if no path exists. Walk through your approach and its time complexity.
  • After finding the shortest path cost, how would you reconstruct the actual path taken?

“Jumped straight to Dijkstra which was right, but I fumbled explaining why BFS alone wouldn't work here.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jun 2026
Amazon SWE coding round, got a grid traversal problem. Pretty standard backtracking territory but the details matter more than you'd expect.
  • Given a 2D grid of characters and a target word, determine whether the word can be found by moving through adjacent cells (up, down, left, right) without reusing any cell.

“I jumped to DFS pretty fast which was the right call, but I fumbled the visited-cell logic at first.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jun 2026
Applied Scientist loop at Amazon, science breadth round covering ML system design with a focus on search and ranking, plus some experimentation case work and behavioral questions. Pretty sparse on details but enough to get a sense of the shape of it.
  • Design an ML system for a search or ranking use case.
  • Walk through how you would design or evaluate an experiment for a ranking or search system.

“This is the core of the round apparently.”

View Post

Amazon·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Amazon system design round for a software engineer role, one long question about building a memory-based feature switcher. The depth they expected was pretty intense and I felt like I was playing catch-up the whole time.
  • Design an in-memory switcher component that toggles between ENABLED and DISABLED states based on runtime memory usage, with user-configurable thresholds, hysteresis to prevent flapping, thread-safe access, runtime reconfiguration, callbacks on state change, and defined behavior under spikes, polling failures, and multi-instance deployments.

“This was basically seven questions dressed up as one.”

View Post

Amazon·Software Engineer·Technical Phone Screen

Jun 2026
Amazon SWE coding round, one question the whole time: number of islands. Seemed straightforward but they kept pushing on scalability which is where things got interesting.
  • Given a 2D binary grid of '1's (land) and '0's (water), count the number of islands, where an island is a group of adjacent land cells connected horizontally or vertically.

“I went with DFS pretty quickly, marking visited cells in place to avoid extra space.”

View Post

Amazon·Research Scientist·Onsite - Multi Round

Jun 2026
Applied Scientist loop at Amazon with a round built entirely around an ambiguous, under-defined business problem. No coding, no system design slides, just a vague prompt and a lot of silence while you figure out what to even ask.
  • Here is a vague business problem. Walk me through how you would approach it.
  • How would you define success for this problem, and what metrics would you use to measure it?
  • What are two or three plausible hypotheses for why this problem exists, and how would you distinguish between them experimentally?
  • If you ran an experiment and saw improvement on your offline metric but no lift in the online A/B test, what would you do?

“The silence after the prompt is genuinely uncomfortable.”

View Post