Senior AI Compiler Architect — MLIR & GPU Performance

NVIDIA Gruppe; Santa Clara (CA); On-site; USD 152,000 - 288,000; Full time

Job summary

NVIDIA is hiring a Senior AI Compiler Engineer to design and optimize an MLIR-based AI compiler powering NVIDIA’s inference engine. You’ll work on graph representations, APIs, and dialect definitions for future GPU architectures, focusing on performance, fast builds, and memory efficiency.

You will collaborate with framework and hardware teams, lead your development efforts, and contribute to kernel generation for neural networks, with opportunities to impact both data center and edge

Qualifications

• Bachelor's, Master's, or Ph.D. in CS/CE or related field, or equivalent experience.
• 3+ years in performance analysis and compiler optimizations.
• Experience with MLIR, XLA, and LLVM.
• Excellent C/C++ and Python programming and design skills.

Responsibilities

• Develop MLIR-based graph representations and optimizations for future GPU architectures.
• Partner with framework and hardware teams to enable new model patterns and upcoming GPU architectural features.
• Define APIs and MLIR dialects, conduct performance optimizations and analysis, implement compiler optimizations and kernel generation for neural networks, and contribute to other general software engineering work.

Skills

C/C++

Python

Performance analysis

Compiler optimizations

MLIR/XLA/LLVM

Independent work

Interpersonal skills

Education

Bachelor's/Master's/PhD in CS or related field

Tools

MLIR

LLVM

CUDA

OpenCL

Job description

NVIDIA is hiring a Senior AI Compiler Engineer to design and optimize an MLIR-based AI compiler powering NVIDIA’s inference engine. You’ll work on graph representations, APIs, and dialect definitions for future GPU architectures, focusing on performance, fast builds, and memory efficiency.

You will collaborate with framework and hardware teams, lead your development efforts, and contribute to kernel generation for neural networks, with opportunities to impact both data center and edge