NVIDIA
Raytracing Compiler Engineer - Developer and Performance Technology
$224K–$431K
US, CA, Santa Clara +5 · Full-time · Posted Aug 5, 2026
- Location
- US, CA, Santa Clara
US, TX, Remote
US, OR, Remote
US, MO, St. Louis
US, UT, Salt Lake City
US, MA, Remote - Experience
- 12+ years of compiler development, including 3+ years of LLVM experience focused on code generation. Preferred: Master’s or PhD with 5+ years of relevant experience.
- Education
- B.S. or higher degree in Computer Science/Engineering or equivalent experience.
About the job
Design and develop an LLVM-based JIT compiler backend for NVIDIA GPUs, including code-generation passes and optimizations for performance and fast compilation.
What you'll do
- Guide design and implementation of the LLVM-based GPU backend JIT compiler.
- Develop compiler passes and optimizations, and analyze performance.
- Coordinate with compiler, hardware, and application teams on improvements and problem resolution.
What they ask for
- B.S. or higher in Computer Science/Engineering or equivalent experience.
- 12+ years of compiler development, including 3+ years of LLVM experience focused on code generation.
- Expertise in LLVM code generation, including instruction scheduling, software pipelining, register allocation, and LLVM IR.
- Proficiency in compiler design and performance analysis tools; strong hands-on C++ skills.
Nice to have
- Master’s or PhD, with 5+ years of relevant experience.
- Experience developing GPU compiler optimizations and code generation, or developing CUDA, DirectX, OpenGL/Vulkan, or OptiX applications.
Pay
224,000 USD - 356,500 USD base salary (Level 5); 272,000 USD - 431,250 USD base salary (Level 6)
Tools
LLVM, C++, CUDA, DirectX, OpenGL/Vulkan, OptiX
AI focus
NVIDIA uses AI tools in its recruiting processes; the role description does not specify AI development work.
Model-written summary · the original posting is authoritative