Senior Staff AI Accelerator Performance Architect

What you’d do

Analyze performance of transformer inference and training workloads on wafer-scale AI hardware, covering attention, GEMM/GEMV, collective communication, mixture-of-experts, quantization, and memory-constrained execution. Also works on kernel optimization, compiler performance, runtime scheduling, and distributed accelerator systems. Full-time in Sunnyvale, CA. Found on the employer's careers page on Oct 6, 2026. Apply through the link.

You’ll need

• Performance analysis of transformer inference or training workloads
• Attention, GEMM/GEMV, collective communication, MoE, quantization
• Kernel optimization, compiler performance, or runtime scheduling
• Distributed accelerator systems experience