NVIDIA
Principal Architect, Accelerated Computing
$272K–$431K
US, CA, Santa Clara · Principal · Full-time · Posted Oct 20, 2026
- Experience
- 12+ years of relevant work experience
- Education
- MS or Ph.D. in Computer Science, Computer Engineering, or Electrical Engineering, or equivalent experience
About the job
Architect and optimize algorithms, software, and system designs for accelerated computing on current and next-generation NVIDIA GPUs.
What you'll do
- Analyze and optimize GPU performance; develop parallel algorithms, data structures, and reference code.
- Assess hardware-software interactions and collaborate across teams to guide accelerated-computing direction and co-design.
- Present and publish work through white papers, conference publications, blog posts, and patent applications.
What they ask for
- MS or Ph.D. in Computer Science, Computer Engineering, or Electrical Engineering, or equivalent experience.
- Strong mathematical fundamentals, including linear algebra and numerical methods.
- Hands-on experience with massively parallel GPU programming, such as CUDA or OpenCL; strong knowledge of C and C++.
- Experience benchmarking, profiling, and characterizing workloads on GPU and CPU clusters.
Nice to have
- Familiarity with MPI or OpenSHMEM/NVSHMEM; threading and Unix-style IPC APIs.
- Familiarity with Python.
Pay
272,000 USD - 431,250 USD (base salary)
Tools
NVIDIA GPUs, CUDA, OpenCL, MPI, OpenSHMEM/NVSHMEM, C, C++, Python
AI focus
The role covers accelerated computing for machine learning and AI, including optimizing algorithms and applications for GPUs.
Model-written summary · the original posting is authoritative