Senior CUDA Kernel Engineer - Tensor Core GEMM

NVIDIA; Durham (NC); On-site; USD 152,000 - 241,500; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking talented professionals in Durham, NC to develop and optimize Tensor Core-based deep learning kernels. As part of the CUTLASS team, you'll be writing and optimizing code for peak performance on NVIDIA architectures.

The ideal candidate has advanced degrees in Computer Science or related fields, strong C++ skills, and relevant industry experience. NVIDIA offers a competitive salary ranging from 152,000 to 287,500 USD based on level and experience.

Apply by June 5, 2026 to join a leading technology employer committed to diversity and inclusion.

Qualifications

• 3+ years of relevant industry experience.
• Strong proficiency in C++ programming and software design.
• Deep understanding of computer architecture.

Responsibilities

• Write Tensor Core-based deep learning kernels using CUTLASS.
• Optimize kernels for peak throughput on software performance simulators.
• Collaborate with teams across NVIDIA for timely kernel delivery.

Skills

C++ programming

CUDA

Math kernels optimization

Software design

Education

Masters or PhD in Computer Science or related field

Tools

CUDA

OpenCL

HIP

Job description

NVIDIA is seeking talented professionals in Durham, NC to develop and optimize Tensor Core-based deep learning kernels. As part of the CUTLASS team, you'll be writing and optimizing code for peak performance on NVIDIA architectures.

The ideal candidate has advanced degrees in Computer Science or related fields, strong C++ skills, and relevant industry experience. NVIDIA offers a competitive salary ranging from 152,000 to 287,500 USD based on level and experience.

Apply by June 5, 2026 to join a leading technology employer committed to diversity and inclusion.