Senior CUDA Kernel Engineer for Exascale AI Compute

SpaceX Super Intelligence; Palo Alto (CA); On-site; USD 180,000 - 440,000; Full time

Benefits offered by this job

Equity

Medical coverage

Vision coverage

Dental coverage

401(k)

Job summary

SpaceXSI’s Compute Infrastructure team is building one of the world’s largest AI supercomputers and you will own the raw GPU hardware and the platform layer that runs on top of it in Palo Alto. You will work across the full stack—from CUDA kernel optimizations to orchestration—to accelerate Grok’s training and inference.

This broad role combines hardcore supercompute with compute infrastructure work and will directly impact Grok’s AI progress by delivering scalable performance and reliability

Qualifications

• Deep low-level systems programming with C/C++/PTX/SASS.
• Experience with large-scale GPU clusters in production.
• Hands-on GPU kernel optimization using CUTLASS and Nsight.
• Track record building high-performance AI training/inference platforms.
• Ability to reason from first principles to optimize memory and compute.

Responsibilities

• Design, build, and optimize massive GPU clusters for extreme-scale training and inference workloads
• Develop and tune low-level CUDA kernels (GeMM, Attention, etc.), using CUTLASS, Tensor Cores, and Nsight for maximum performance
• Profile, debug, and eliminate bottlenecks across GPU memory hierarchy, networking fabric, filesystems, and multi-GPU operation
• Collaborate closely with AI research teams to deliver production-grade performance and scalability

Skills

Low-level systems

GPU clusters

CUDA kernels

Nsight profiling

AI infra experience

Job description

SpaceXSI’s Compute Infrastructure team is building one of the world’s largest AI supercomputers and you will own the raw GPU hardware and the platform layer that runs on top of it in Palo Alto. You will work across the full stack—from CUDA kernel optimizations to orchestration—to accelerate Grok’s training and inference.

This broad role combines hardcore supercompute with compute infrastructure work and will directly impact Grok’s AI progress by delivering scalable performance and reliability