CUDA DL Systems Engineer — Equity Options & HPC Performance

Nvidia Corporation in; Santa Clara (CA); On-site; USD 124,000 - 196,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA in Santa Clara is seeking a Software Engineer for CUDA Deep Learning Systems (Finance) to push the performance frontier of AI workloads, from a single GPU to large clusters.

You will design high-performance CUDA kernels, prototype distributed systems, and collaborate with AI researchers, HW/SW architects, and driver experts to scale multi-node training and inference.

This role offers equity and benefits and is integral to advancing next-generation accelerator technologies.

Qualifications

• BS, MS, or PhD in Computer Science, Computer Engineering, Electrical Engineering, or related field (or equivalent experience).
• 2+ years of relevant industry or academic experience after degree.
• Strong proficiency in C++ and Python programming.
• Solid background in Deep Learning fundamentals with a focus on transformers.
• Understanding of distributed computing and multi-node scaling.
• Experience profiling and optimizing GPU-based workloads with CUDA.

Responsibilities

• Explore, research, and prototype novel systems optimizations for advanced deep learning models at the intersection of high-level DL frameworks and CUDA through modeling, simulation, and silicon prototyping.
• Architect and optimize distributed computing systems that scale from a single node to cluster-scale environments.
• Design, implement, and optimize custom high-performance CUDA kernels for evolving neural network architectures.
• Analyze hardware-software interactions to identify and resolve performance bottlenecks in training and inference.
• Collaborate with AI researchers, HW/SW architects, kernel and compiler authors, and CUDA driver experts to improve accelerator utilization and memory bandwidth.
• Develop exploratory tools and runtime systems to profile and accelerate new paradigms in deep learning.
• Write clean, maintainable code for prototypes that can transition to open-source releases or internal tools.

Skills

C++

Python

Deep Learning basics

Distributed computing

CUDA programming

Kernel optimization

Performance profiling

Research experience

Team collaboration

Education

BS/MS/PhD in CS/CE/EE

Tools

CUDA

NCCL

MPI

UCX

Triton

XLA

Torch.compile

Job description

NVIDIA in Santa Clara is seeking a Software Engineer for CUDA Deep Learning Systems (Finance) to push the performance frontier of AI workloads, from a single GPU to large clusters.

You will design high-performance CUDA kernels, prototype distributed systems, and collaborate with AI researchers, HW/SW architects, and driver experts to scale multi-node training and inference.

This role offers equity and benefits and is integral to advancing next-generation accelerator technologies.