Senior DL Systems Architect for Next‑Gen AI GPUs

NVIDIA Corporation; Santa Clara (CA); On-site; USD 224,000 - 357,000; Full time

Job summary

NVIDIA Corporation seeks a senior system architect to advance AI system design and research collaboration. You will contribute to features for next-generation GPUs, map AI workloads to hardware, and prototype optimized architectures.

You will stay current with deep learning research and work with diverse teams including researchers, hardware architects, and engineers. The role emphasizes solving bottlenecks through first-principles analyses, benchmarking, and co-design from conception to

Qualifications

• DL & ML fundamentals required across DL workloads.
• Experience adapting/training DNNs for tasks.
• Proficiency with major DL frameworks (PyTorch, TensorFlow, JAX).
• Numerical/ performance analysis and architectural optimization experience.
• Programming fluency in C++ and Python.
• GPU computing experience (CUDA/OpenCL/OpenACC) and HPC (MPI/OpenMP).

Responsibilities

• Contribute to features for next-generation GPUs and AI systems.
• Keep up with DL research; collaborate with diverse teams.
• Co-design architecture for systems from conception to prototyping.
• Map AI/DL workloads to hardware and system capabilities.
• Identify bottlenecks and propose solutions to improve systems.
• Perform first-principles analyses and build prototypes; benchmark ideas.

Skills

DL & ML fundamentals

DNN training adaptation

C++ programming

Python programming

GPU computing (CUDA/OpenCL/OpenACC)

HPC (MPI/OpenMP)

DL frameworks (PyTorch/TF/JAX)

Numerical & performance analysis

Computer architecture

Education

MS or PhD in CS/CompArch/EE

Tools

CUDA

OpenCL

OpenACC

MPI

OpenMP

PyTorch

TensorFlow

JAX

Python

C++

Job description

NVIDIA Corporation seeks a senior system architect to advance AI system design and research collaboration. You will contribute to features for next-generation GPUs, map AI workloads to hardware, and prototype optimized architectures.

You will stay current with deep learning research and work with diverse teams including researchers, hardware architects, and engineers. The role emphasizes solving bottlenecks through first-principles analyses, benchmarking, and co-design from conception to