Amazon Machine Learning Engineer Interview Questions
Master your tech interview with our curated database of real questions from top companies.
Explain key ML theory and techniques
Onsite Machine Learning Engineer: Mixed Topics You are asked to answer concisely but with depth across the following topics: 1) XGBoost Parallel Compu...
Explain ML statistics and model design concepts
Technical Phone Screen: Theory + System Design Probability and Statistics 1. Define a moment generating function (MGF) and explain how it is used. 2. ...
Explain parallelism and collectives in training
Parallelism strategies and communication in large-scale training You are designing a distributed training setup for very large neural networks that ca...
Compare float types and design ablation
Floating-point types and ablation study design You are training deep neural networks on modern accelerators that support multiple floating-point forma...
Implement PyTorch training loop
Implement a basic PyTorch training loop You are given a PyTorch neural network model, a DataLoader that yields (inputs, targets) batches, an optimizer...
Explain weight initialization methods and goals
Explain why weight initialization matters in deep neural networks. Then describe common initialization methods (such as random normal/uniform, Xavier/...
List hyperparameter tuning methods
Describe common methods for hyperparameter tuning in machine learning. For each method, explain: - How it works conceptually. - Its advantages and dis...
Contrast CNNs and fully connected networks
Compare convolutional neural networks (CNNs) with fully connected (dense) networks. Explain: - The structural differences between convolutional layers...
Analyze attention complexity and improvements
In the context of Transformer-style models, analyze the computational complexity of self-attention. Assume a sequence length of \(n\) and hidden dimen...
Compare decision trees and random forests
Compare decision trees and random forests. In your answer, discuss: - How a single decision tree is built and its main advantages and disadvantages. -...
Explain vanishing gradients and activations
Explain the vanishing gradient problem in deep neural networks. In your answer: - Describe how backpropagation works at a high level and why gradients...
Describe overfitting and L1/L2 regularization
Define overfitting in machine learning and explain why it is harmful. Then describe L1 and L2 regularization: - How each one modifies the loss functio...
Explain the bias–variance trade-off
Explain the bias–variance trade-off in supervised learning. In your answer, cover: - What bias and variance mean in the context of a prediction model....
Design an S3-like object storage service
Design a cloud object storage service similar to Amazon S3. The service should allow clients to upload, store, and download large files reliably and e...
Find shortest path in a grid with obstacles
You are given a 2D grid of size m x n representing a maze. Each cell in the grid is either empty (0) or blocked (1). You are also given two coordinate...
Design an LLM quality validation system
You are asked to design an end-to-end LLM quality validation system for a team that trains and serves large language models. The goal is to automatica...
Implement integer division without using division
You are given two 32-bit signed integers dividend and divisor. Implement a function that divides dividend by divisor and returns the integer quotient,...
Explain Transformers and MoE in LLMs
You are interviewing for a role working with large language models (LLMs). Explain the following concepts and how they relate to building and scaling ...
Design logo infringement detection system
Scenario You work for a large e-commerce company. Brands register their official logos with you (e.g., Nike swoosh, Apple logo, etc.). Third-party sel...
Design LFU cache with distributed extension
Problem You are asked to design and implement a data structure that behaves like an in-memory cache with a Least Frequently Used (LFU) eviction policy...