Software Engineer - Kernels/CUDA (C++)

About this role

xAI seeks a Software Engineer to optimize and build a massive GPU supercomputer infrastructure for AI training and inference. You'll work across the full stack—from low-level CUDA kernel optimization to distributed cluster orchestration—directly accelerating Grok's performance and scalability.

What you'll do

• Design and optimize massive GPU clusters for extreme-scale AI training and inference
• Develop and tune low-level CUDA kernels (GeMM, Attention) using CUTLASS and Tensor Cores
• Work on Linux kernel internals, scheduling, memory management, and resource isolation
• Build custom container orchestration and virtualization layers beyond standard Kubernetes
• Profile and eliminate bottlenecks across GPU memory, networking, filesystems, and multi-GPU operations
• Create infrastructure-as-code, automation, and tools to maintain supercomputer reliability

What they're looking for

• C/C++ or Rust systems programming
• GPU kernel optimization and CUTLASS
• Large-scale GPU cluster operations
• Linux kernel internals and virtualization
• Distributed systems and orchestration
• High-performance storage systems
• Performance profiling and debugging
• First-principles optimization reasoning

Benefits

• Competitive salary $180,000–$440,000 USD
• Equity compensation
• Comprehensive medical, vision, and dental coverage
• 401(k) retirement plan
• Short and long-term disability insurance
• Life insurance