6 days ago
Base Salary
$184k - $357k/yr
Responsibilities
- Train, develop, and deploy generative AI models including LLMs and diffusion models using NVIDIA’s AI software stack.
- Build model graph analysis and extraction capabilities using TorchDynamo, torch.export, and torch.compile.
- Develop inference optimization techniques including automated model sharding, tensor parallelism, sequence parallelism, and efficient attention kernels with KV caching.
- Analyze and profile GPU kernel-level performance to identify hardware and software optimization opportunities.
- Architect and design a modular, scalable software platform supporting broad model coverage and optimization techniques.
- Collaborate with NVIDIA teams to integrate high-performance kernel implementations into automated deployment solutions.
Requirements
- Master’s degree, PhD, or equivalent experience in Computer Science, AI, Applied Math, or a related field.
- 8+ years of relevant work or research experience in deep learning.
- Excellent software design, debugging, performance analysis, and test design skills.
- Strong proficiency in Python, PyTorch, and related machine learning tools such as HuggingFace.
- Strong algorithms and programming fundamentals.
- Good written and verbal communication skills and the ability to work independently and collaboratively.
- Contributions to PyTorch, JAX, or other machine learning frameworks are advantageous.
- Knowledge of GPU architecture and compilation stacks, including the ability to understand and debug end-to-end performance, is advantageous.
- Familiarity with NVIDIA deep learning SDKs such as TensorRT is advantageous.
- Experience writing high-performance GPU kernels using CUDA, CUTLASS, or Triton is advantageous.
Benefits
- Eligible for equity and comprehensive benefits.
- Applications will be accepted at least until October 13, 2026.
- The posting is for an existing vacancy.
Categories
About Nvidia
Nvidia designs and sells GPUs and accelerated computing platforms for data centers, AI/ML, graphics, gaming, and automotive, monetizing through hardware, software platforms (CUDA, AI frameworks), and systems like DGX and networking. Customers include cloud providers, enterprises, researchers, and OEMs. Founded in 1993 and headquartered in Santa Clara, it is a public company traded on NASDAQ under NVDA.