Lead HPC Library Engineer: CUDA-Accelerated AI

NVIDIA; United States; Remote; USD 140,000 - 230,000; Full time

Job summary

NVIDIA is seeking a highly skilled software architect to design, develop, and optimize GPU-accelerated data processing libraries. You will work on projects like DALI and nvImageCodec, shaping performance and scalability across platforms.

The role requires 5+ years in high-performance parallel numerical software with CUDA, strong C/C++ skills, and proven project leadership. Collaboration with cross-functional teams and a focus on user experience are essential.

Qualifications

• PhD or MSc in Computational Science, Computer Science, Applied Math, or related field is preferred (or equivalent experience).
• 5+ years experience developing, debugging, and optimizing high-performance parallel numerical applications with GPU acceleration using CUDA.
• Proficient in C/C++ programming and software development.

Responsibilities

• Develop scalable library software using modern tools and languages for numerical methods.
• Performance tuning, optimization, and benchmarking of algorithms on various architectures.
• Collaborate with leadership and partners to define feature and performance requirements and contribute to roadmaps.
• Provide technical leadership and guidance to library engineers.
• Identify opportunities to improve user experience and library performance.

Skills

CUDA

C/C++

Parallel computing

Leadership

Education

PhD or MSc in Computational Science/CS/Applied Math

Tools

JIRA

CI/CD

Job description

NVIDIA is seeking a highly skilled software architect to design, develop, and optimize GPU-accelerated data processing libraries. You will work on projects like DALI and nvImageCodec, shaping performance and scalability across platforms.

The role requires 5+ years in high-performance parallel numerical software with CUDA, strong C/C++ skills, and proven project leadership. Collaboration with cross-functional teams and a focus on user experience are essential.