Member of Technical Staff, Kernels

The role

You will design, implement, and maintain high-performance kernels to optimize throughput and latency during training and inference. The work focuses on long-context windows and memory utilization.

What you will do

• Design and implement kernels that support high-performance long-context behavior
• Ownership of kernel design, implementation, deployment, and production reliability
• Focus on robustness, extensive testing, and functional correctness
• Evaluate porting compute kernels to alternative hardware options
• Co-design kernels with training, inference, and RL teams

Required qualifications

• Low-level programming experience targeting AI accelerators such as NVIDIA Blackwell or Google TPUs
• Develop and optimize GPU kernels in frameworks such as NCCL, MSCCLPP, CUTLASS, CuTeDSL, Triton, Quack, Flash-Attention, and similar frameworks
• Agility, ownership mindset, and grit

Nice to have

• Experience in other kernel authoring frameworks such as Pallas or Mojo

Benefits

• 401(k) plan with 6% salary matching
• Generous health, dental and vision insurance for you and your dependents
• Unlimited paid time off
• Relocation stipend to bring you to SF, if possible

Before you apply

• Visa sponsorship available to bring you to SF, if possible

About the company

Magic's mission is to build safe AGI that accelerates humanity's progress. The approach combines frontier-scale pretraining, domain-specific RL, ultra-long context, and inference-time compute.