Performance Engineer in ML Kernel Design

Embrace the challenge at Amazon Devices as a Performance Engineer specializing in ML Kernel design for Edge AI. Focus on developing robust CUDA and Triton kernels that optimize neural network compression.

This position requires you to collaborate closely with various teams, designing kernel-level solutions that streamline model training and inference processes. Your technical expertise will help propel Amazon's advanced compression platform, achieving unrivaled efficiency and performance across AI applications.

Key Responsibilities:
• Implement CUDA and Triton kernels for efficient ML training
• Conduct performance analysis and resolve critical bottlenecks
• Design and enhance kernel optimization techniques
• Support team members with kernel development tools
• Maintain libraries for training kernel improvements

Requirements:
• 3+ years' experience in software development
• 2+ years in system design and ar...