GPGPU Software Architect: Lead CUDA Stack & AI Acceleration

XPENG; Santa Clara (CA); On-site; USD 241,800 - 409,200; Full time

Benefits offered by this job

Medical insurance

Vision insurance

401(k)

Job summary

A leading smart technology company is seeking a GPGPU Software Architect in Santa Clara, CA. The role focuses on developing a comprehensive software stack compatible with CUDA and collaborating with cross-functional teams in AI and GPU technologies. Candidates should have over ten years of experience in systems software with proficiency in CUDA. Benefits include competitive salary, medical insurance, and 401(k).

Qualifications

• 10+ years in systems software, with 5+ years in designing CUDA Compute stacks.
• Experience leading end-to-end development of GPU Runtime or AI acceleration library.
• Comprehensive mastery of PTX/SASS, CUDA Driver API, and cuBLAS/cuDNN internals.

Responsibilities

• Develop and refine a roadmap for a software stack compatible with CUDA.
• Build an observability platform for real-time metrics.
• Collaborate across teams to develop Device Plugins and GPU Operators.

Skills

Software Technical Strategy

CUDA expertise

AI framework collaboration

Benchmarking and observability

Tools

CUDA

PCIe

LLVM

Job description

A leading smart technology company is seeking a GPGPU Software Architect in Santa Clara, CA. The role focuses on developing a comprehensive software stack compatible with CUDA and collaborating with cross-functional teams in AI and GPU technologies. Candidates should have over ten years of experience in systems software with proficiency in CUDA. Benefits include competitive salary, medical insurance, and 401(k).