Senior Performance Compiler Engineer - Triton
$184k - $288k • US, WA, Redmond
Posted 2d ago
About the job
NVIDIA is seeking a Senior Performance Compiler Engineer to join their open-source Triton compiler project. This role focuses on improving AI performance on NVIDIA GPUs by leveraging compiler technology. You will work on cutting-edge AI applications, including large language models and agents, accelerating both training and inference. The position offers a collaborative and supportive environment to push the boundaries of AI computing.
Responsibilities
- Investigate current and future NVIDIA GPU hardware architecture and programming models.
- Optimize AI performance by understanding advanced algorithms and numerics.
- Design and implement compiler technology using MLIR to generate efficient GPU code from Triton's Python DSL.
- Utilize inline PTX for hand-tuning critical code paths when necessary.
- Engage in iterative optimization processes, starting from kernels or compilers.
- Collaborate with internal NVIDIA teams, including hardware architects and the CUDA compiler team.
Requirements
- Bachelor's, Master's, or Ph.D. degree in Computer Science, Computer Engineering, Applied Math, or related field, or equivalent experience.
- 8+ years of relevant industry experience in software development.
- Strong C++ programming and software design skills, with a focus on performance analysis and debugging.
- Experience in parallel programming, including CUDA/OpenCL GPU programming or OpenMP.
- Solid understanding of computer architecture and hands-on experience with assembly-level programming.
Benefits
- Competitive salaries
- Generous benefits package
- Equity