GPU TensorIR & CUDA Tile Compiler Engineer

NVIDIA AI; Redmond (WA); On-site; USD 152,000 - 242,000; Full time

Benefits offered by this job

Health benefits

Job summary

NVIDIA is seeking a software engineer to work on CUDA Tile and TensorIR compiler technologies. You will design and implement compiler transformations, MLIR-based dialects, and lowering passes to optimize tile-based kernels on NVIDIA GPUs.

The role requires strong C/C++ programming skills, 3+ years in compiler optimization, and the ability to work independently within a dynamic team. Equity and benefits accompany the base salary.

Qualifications

• Bachelor’s, Master’s or Ph.D. in Computer Science, Computer Engineering, or related field (or equivalent).
• 3+ years of experience in compiler optimization, performance analysis and IR design.
• Ability to work independently, define goals and lead development.
• Excellent C/C++ programming and software design, debugging and testing.

Responsibilities

• Design and implement compiler transformations and MLIR-based dialects for CUDA Tile and TensorIR.
• Optimize performance of tile-based kernels across NVIDIA GPU architectures.
• Define public APIs and contribute to compiler and optimization techniques.

Skills

C/C++ programming

Compiler optimization

IR design

Performance analysis

Independent work

Interpersonal skills

Education

Bachelor's/Masters/Ph.D. in CS/CE or related

Tools

MLIR

LLVM

XLA

TVM

Job description

NVIDIA is seeking a software engineer to work on CUDA Tile and TensorIR compiler technologies. You will design and implement compiler transformations, MLIR-based dialects, and lowering passes to optimize tile-based kernels on NVIDIA GPUs.

The role requires strong C/C++ programming skills, 3+ years in compiler optimization, and the ability to work independently within a dynamic team. Equity and benefits accompany the base salary.