Implement Multi-Head Self-Attention

Quick Overview

This question evaluates understanding of multi-head self-attention and the competency to implement transformer attention modules using learned Q/K/V projections, head-wise tensor reshaping, attention masking, and PyTorch tensor operations.

Implement Multi-Head Self-Attention

Company: Uber

Role: Machine Learning Engineer

Category: Machine Learning

Difficulty: medium

Interview Round: Technical Screen

Quick Answer: This question evaluates understanding of multi-head self-attention and the competency to implement transformer attention modules using learned Q/K/V projections, head-wise tensor reshaping, attention masking, and PyTorch tensor operations.

|Home/Machine Learning/Uber
Uber logo
Uber
Jan 10, 2026, 12:00 AM
mediumMachine Learning EngineerTechnical ScreenMachine Learning
14
0
Loading...
Loading comments...