Machine Learning Engineer — GPU Kernel

Job overview

The Machine Learning Engineer — GPU Kernel is an onsite role in Sunnyvale, CA, with compensation of $150,000–$450,000 per year. The Institute of Foundation Models is a research lab focused on building, understanding, using, and risk-managing foundation models; this role supports its high-performance computing and deep learning work.

What you'll do

• Analyze, profile, and optimize deep learning workloads
• design performance benchmarks and testing methods
• build tools for workload analysis and optimization
• investigate system bottlenecks and improve GPU utilization
• develop kernels and systems for new model architectures
• participate in design reviews and code reviews
• document technical work
• and validate CUDA kernel outputs and gradients against reference implementations.

What we're looking for

Skills & competencies

machine learning engineer

gpu kernel

onsite

sunnyvale ca

institute of foundation models

cuda

c++

python

pytorch

gpu optimization

deep learning

foundation models

$150k-$450k/year

Work arrangement

• Weekend coverage required

Benefits & perks

• Medical, dental, and vision benefits
• bonus
• 401(k)
• paid time off, sick leave, and holidays
• paid parental leave
• employee assistance program
• life insurance and disability.

Why this role

The role focuses on CUDA kernel development and optimization. Required skills include C++, CUDA, GPU profiling, Python, and integration with PyTorch or an equivalent framework. Triton, CUTLASS, PTX/SASS analysis, distributed training, and mixed-precision validation are listed as nice-to-haves.

About the employer

Institute of Foundation Models is hiring for this role. Industry: Research and Development in the Social Sciences and Humanities. Sector: 54.