Senior Rust Engineer for CUDA Core Libraries

Rustify; United States; Hybrid; USD 180,000 - 240,000; Full time

Job summary

NVIDIA is hiring a Senior Software Engineer to develop the Rust experience for CUDA Core Libraries. You will design safe, idiomatic Rust APIs for GPU computing and closely integrate with native C/C++ components.

You will own features end-to-end, optimize cross-boundary performance, and collaborate with engineers across Rust, C/C++, Python, and runtime teams to deliver reliable, high-performance CUDA libraries.

Qualifications

• Experience designing safe, idiomatic Rust APIs for complex systems.
• Strong understanding of interoperating Rust with C/C++ and CUDA interfaces.
• Experience with testing, profiling, and benchmarking GPU-capable software.

Responsibilities

• Design and implement idiomatic Rust libraries and APIs for foundational CUDA functionality and GPU algorithms.
• Build safe Rust abstractions over native CUDA and C/C++ interfaces.
• Develop and maintain the C/C++ components required to support Rust-facing functionality.
• Establish safe and efficient interoperability between Rust and C/C++, including interfaces that support downstream Python integration.
• Optimize performance across Rust, native C/C++, and GPU execution boundaries.
• Own features end-to-end: design, implementation, testing, profiling, benchmarking, documentation, packaging, release, and maintenance.
• Improve the Rust developer experience through examples, diagnostics, build integration, compatibility testing, CI, and collaboration with C/C++, Python, compiler, and runtime engineers.
• Work with users to investigate and resolve issues related to safety, correctness, usability, compatibility, and performance.

Skills

Rust

C/C++ interoperability

API design

Python integration

Performance optimization

Education

BS/MS/PhD in CS/CE or related field

Job description

NVIDIA is hiring a Senior Software Engineer to develop the Rust experience for CUDA Core Libraries. You will design safe, idiomatic Rust APIs for GPU computing and closely integrate with native C/C++ components.

You will own features end-to-end, optimize cross-boundary performance, and collaborate with engineers across Rust, C/C++, Python, and runtime teams to deliver reliable, high-performance CUDA libraries.