Senior Software Engineer, CUTLASS Performance

at Nvidia

USD 152,000-287,500 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI Deep Learning GPU HPC JAX @ 4 LLM @ 4 Performance Analysis @ 4 Performance Optimization @ 4 PyTorch @ 4 Python @ 7 SGLang @ 4 vLLM @ 4

Details

NVIDIA's high-performance computing platforms are powering the AI revolution across many applications and industries. Within our software stack, CUTLASS stands out as a popular open-source ecosystem dedicated to high-performance linear algebra and Tensor Core primitives. Since 2017, it has provided the community with C++ and Python abstractions to implement custom matrix multiply (GEMM) and related math and deep learning computations on NVIDIA GPUs.

If you are enthusiastic about performance and eager to help bridge the gap between current performance and what’s theoretically possible, apply to join the CUTLASS team today!

Responsibilities

Benchmark the performance of state-of-the-art deep learning models’ inference and training passes to identify key GPU kernel and fusion opportunities.
Identify gaps between theoretical and realized performance, and suggest software improvements or model adjustments to resolve them.
Develop tooling to automate the benchmarking, analysis, and performance optimization loop to push the limit of CUTLASS kernel performance within DL networks.
Be the authoritative resource on kernel performance in the team and engage with teams across NVIDIA including GPU architecture, DL frameworks, and QA as the performance representative for the CUTLASS team.

Requirements

Masters or PhD degree in Computer Science, Computer Engineering, or related field (or equivalent experience).
3+ years of relevant industry experience.
Strong programming skills in Python and C++.
Experience in software performance analysis and optimization.
Deep understanding of computer architecture and familiarity with GPUs or similar parallel processing architectures.

Ways to stand out from the crowd

Deep understanding of state-of-the art DL model architectures.
Hands-on experience with performance benchmarking of DL frameworks like PyTorch, JAX, SGLang, vLLM, TRT-LLM, or others.
Experience in developing performance models and performance regression systems.

Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 152,000 USD - 241,500 USD for Level 3, and 184,000 USD - 287,500 USD for Level 4.