Found Description
Drive AI innovation at Amazon Devices as a Performance Engineer specializing in ML Kernel development in Vancouver, British Columbia. Focus on high-performance CUDA and Triton kernels for cutting-edge compression technology.
Join the AI Platform team at Amazon Devices, where you will engineer the next generation of edge AI capabilities. This role emphasizes your expertise in kernel performance optimization to boost training speed significantly. Work closely with Applied Scientists and hardware architects as you implement advanced techniques for model compression and inference on state-of-the-art hardware.
Key Responsibilities:
• Design high-performance CUDA and Triton kernels for quantization training
• Analyze kernel performance and resolve bottlenecks through profiling
• Implement operator fusion and memory optimization techniques
• Build profiling harness for kernel performance testing
• Collaborate on ML solutions with scientists and hardware architects
Requ...
Join the AI Platform team at Amazon Devices, where you will engineer the next generation of edge AI capabilities. This role emphasizes your expertise in kernel performance optimization to boost training speed significantly. Work closely with Applied Scientists and hardware architects as you implement advanced techniques for model compression and inference on state-of-the-art hardware.
Key Responsibilities:
• Design high-performance CUDA and Triton kernels for quantization training
• Analyze kernel performance and resolve bottlenecks through profiling
• Implement operator fusion and memory optimization techniques
• Build profiling harness for kernel performance testing
• Collaborate on ML solutions with scientists and hardware architects
Requ...
Ready to Apply?
Submit your application for Performance Engineer for ML Kernel at Amazon at Amazon
Apply Now