Senior Performance Compiler Engineer - Triton

Job Title

Senior Performance Compiler Engineer - Triton

Role Summary

Join the Triton compiler team to design and implement compiler technology that improves AI training and inference performance on NVIDIA GPUs. The role focuses on using MLIR and Triton's Python DSL to generate highly efficient GPU code and collaborating with hardware architects and compiler teams to influence future products.

This is an engineering position centered on performance optimization, low-level code generation, and cross-team technical collaboration.

Experience Level

Senior - typically 8+ years of relevant industry software development experience.

Responsibilities

Key responsibilities include researching GPU hardware and programming models, designing compiler optimizations, and extracting peak performance from critical kernels.

• Investigate current and future NVIDIA GPU architectures and programming models to find optimization opportunities.
• Analyze AI algorithms and numerics (e.g., attention mechanisms, MoEs, block-scaled floating point) to guide optimizations.
• Design and implement compiler passes using MLIR to lower Triton Python DSL kernels to efficient GPU code.
• Hand-tune performance-critical paths with inline PTX or assembly when necessary.
• Iteratively profile and optimize at both kernel and compiler levels to achieve peak performance.
• Collaborate with hardware architects, CUDA compiler teams, and other internal partners to influence product direction and ensure efficient code generation.

Requirements

Must-have technical skills and experience required for the role.

Must-have:

• 8+ years of professional software development experience.
• Strong C++ programming and software design skills with emphasis on performance analysis and debugging.
• Experience in parallel programming (CUDA/OpenCL) or other parallel models such as OpenMP.
• Solid understanding of computer architecture and hands-on experience with assembly-level programming.

Nice-to-have:

• Experience tuning BLAS or deep learning library kernels.
• Background in numerics and linear algebra relevant to ML workloads.
• Familiarity with MLIR, TVM, or other machine-learning compilers.
• Contributions to open-source projects in AI/ML, compilers, or HPC.
• Awareness of recent research in AI algorithms and numerics.

Education Requirements

Bachelor's, Master's, or Ph.D. in Computer Science, Computer Engineering, Applied Mathematics, or a related technical field - or equivalent practical experience.

About the Company

Company: NVIDIA

Headquarters: Santa Clara, California, USA

NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.