Senior Compiler Engineer - Compute

Details

We are looking for a highly motivated Compiler Software Engineer to join a dynamic and innovative CUDA Compiler team. You will craft compiler capabilities that enable development and deployment of high-performance parallel applications on NVIDIA's next-generation GPUs. The team delivers language extensions and performance improvements to better realize the potential of NVIDIA hardware for a growing range of computational workloads, including deep learning, scientific computation, self-driving cars, video games, and virtual reality.

In this role, as a leader on the team, you will work on the optimizing CUDA C++ compiler, a core component of the platform for accelerating general purpose computation on the GPU. The compiler translates parallel programs written in CUDA/C++ into microcode optimized for fast execution on NVIDIA GPU architectures. You will solve challenging problems alongside experts in GPU computing and systems software.

Responsibilities

• Design and implement enhancements to the CUDA C++ compiler driver and front-end, ensuring tight integration with third-party compiler software.
• Contribute to the CUDA C++ programming language specification and its evolution, including support for new GPU hardware capabilities.
• Drive language conformance, front-end robustness, and compile-time improvements that benefit a wide range of users.
• Provide engineering support for the compiler and deliver targeted changes for production customer releases.
• Establish strong customer and partner relationships through excellent execution and high-quality software development.

Requirements

• BS/MS/PhD in Computer Science, Computer Engineering, or related field (or equivalent experience).
• 3+ years of experience developing production compilers or related software.
• Knowledge of compiler front-end technology and compiler driver development.
• Advanced C++ development skills, including experience working with large, modularized codebases.
• Experience writing and debugging multithreaded programs.
• Strong interpersonal, verbal, and written communication skills; ability to achieve objectives under tight deadlines and manage competing priorities.
• Deep understanding of technology, strong teamwork, and proven ability to guide and influence within a dynamic matrix environment.

Ways to stand out

• Strong track record in compiler engineering with EDG, LLVM-clang, or other industrial-strength compiler front-ends.
• Prior work on domain-specific front-ends or IRs such as MLIR, including dialect development or integration with custom compilers.
• CUDA C++ application development experience.
• Strong understanding of parallel programming concepts.
• Experience working with programming languages and language standards.

Compensation & Benefits

• Base salary range by level:
  • Level 3: 148,000 USD - 235,750 USD
  • Level 4: 184,000 USD - 287,500 USD
• Eligibility for equity and company benefits.

Additional information

• Applications accepted at least until July 31, 2025.
• NVIDIA is an equal opportunity employer committed to diversity and inclusion.