Found Description
Join Amazon Devices as an ML Kernel Performance Engineer in Vancouver, Canada. This role focuses on cutting-edge edge AI technologies, utilizing CUDA and Triton kernels to optimize performance for compression algorithms.
As a key member of the AI Platform team, you will design high-performance CUDA and Triton kernels that enhance GPU execution efficiency. Your expertise will ensure optimal throughput for various AI workloads, translating innovative compression schemes into scalable solutions for edge and cloud environments.
Key Responsibilities:
• Design high-performance CUDA and Triton kernels for edge AI training
• Optimize kernel-level performance for compression training workloads
• Implement advanced operator fusion and memory access optimizations
• Build a kernel development harness for team-wide usability
• Collaborate with Applied Scientists and hardware architects on ML solutions
Requirements:
• 3+ year...
As a key member of the AI Platform team, you will design high-performance CUDA and Triton kernels that enhance GPU execution efficiency. Your expertise will ensure optimal throughput for various AI workloads, translating innovative compression schemes into scalable solutions for edge and cloud environments.
Key Responsibilities:
• Design high-performance CUDA and Triton kernels for edge AI training
• Optimize kernel-level performance for compression training workloads
• Implement advanced operator fusion and memory access optimizations
• Build a kernel development harness for team-wide usability
• Collaborate with Applied Scientists and hardware architects on ML solutions
Requirements:
• 3+ year...
Ready to Apply?
Submit your application for ML Kernel Performance Engineer at Amazon at Amazon
Apply Now