Found Description
Join Amazon Devices as an ML Kernel Performance Engineer in Vancouver, British Columbia, optimizing AI capabilities. Collaborate on high-performance kernel development for neural network compression technology.
In this critical role within the AI Platform team, you will architect CUDA and Triton kernels tailored for advanced model compression and inference tasks. Your efforts will shape the efficiency of AI model training, transforming long processing times into streamlined workflows. Collaborate alongside scientists and engineers to unify software and hardware solutions at scale.
Key Responsibilities: • Implement efficient CUDA and Triton kernels for AI training • Conduct performance analysis to improve kernel efficiency • Optimize memory access patterns for better training throughput • Develop kernel profiling tools to enhance testing efficiency • Maintain library of training kernels with clean interfaces
Requirements: • 3+ years of professional software de...
In this critical role within the AI Platform team, you will architect CUDA and Triton kernels tailored for advanced model compression and inference tasks. Your efforts will shape the efficiency of AI model training, transforming long processing times into streamlined workflows. Collaborate alongside scientists and engineers to unify software and hardware solutions at scale.
Key Responsibilities: • Implement efficient CUDA and Triton kernels for AI training • Conduct performance analysis to improve kernel efficiency • Optimize memory access patterns for better training throughput • Develop kernel profiling tools to enhance testing efficiency • Maintain library of training kernels with clean interfaces
Requirements: • 3+ years of professional software de...
Ready to Apply?
Submit your application for Amazon ML Kernel Performance Engineer Role at Amazon
Apply Now