Overview In this role you will help design hardware accelerators and processor architectures to advance AI on NVIDIA's next gen mobile, embedded, and datacenter platforms. You collaborate with DL researchers, hardware architects, and software engineers to analyze DL methods and propose features that accelerate AI workloads. You will balance performance, power, and scale while staying current with the latest DL research. This is a high-impact opportunity to shape real-time AI computing at a world class, innovation driven company.
Compensation / Benefits- equity
- benefits
- hybrid work option
- competitive base salary (184,000 USD - 287,500 USD)
- career growth
- diverse and inclusive environment
Responsibilities- Analyze deep learning method behavior and performance
- Propose hardware features to accelerate DL workloads (e.g., GPUs, fused operations)
- Study benefits and tradeoffs of proposed features across workloads
- Collaborate with DL researchers, hardware architects, and software engineers
- Contribute to next-generation DL architectures for AI platforms
Key requirements- 5+ years in computer architecture, GPU and system-level design
- Performance analysis and optimization experience
- Experience with LLM workloads and parallelization/fusion strategies
- Experience with core DL kernels (matrix multiply, attention, convolution)
- Programming fluency in C++ (and Python preferred)
- Experience with GPU computing (CUDA)
- Experience with PyTorch
- cross-functional collaboration
- autonomy
- strong problem-solving mindset
- GPU architecture
- CUDA
- PyTorch
