Senior Compiler Optimization Engineer - MLIR

We are seeking an experienced Compiler Optimization Engineer for our Compute Compiler Team at NVIDIA. The successful candidate will work on optimizing CUDA and other compute compilers to maximize the potential of NVIDIA GPUs for various computational workloads, including deep learning and scientific computation.

Key Responsibilities:

• Analyze application code performance on NVIDIA GPUs using profiling tools.
• Identify opportunities to improve performance in the MLIR/LLVM-based compiler middle-end optimizer.
• Develop novel ideas for the compilation pipeline to enhance AI workload performance.
• Design and implement new compiler passes and optimizations for best-in-class performance.
• Collaborate with the Open-source LLVM community for tighter integration.
• Work with distributed teams to oversee improvements and resolve issues.

Requirements:

• B.S., M.S., or Ph.D. in Computer Science, Computer Engineering, or related fields.
• 8+ years of experience in compiler optimizations.
• Excellent C++ programming skills.
• Understanding of processor ISA.
• Strong software engineering principles.

Preferred Qualifications:

• Experience with CUDA or parallel programming languages.
• Knowledge of parallel programming concepts.
• Familiarity with MLIR, LLVM, and Clang compiler development.
• Experience with deep learning frameworks and NVIDIA GPUs.

NVIDIA offers highly competitive salaries and a comprehensive benefits package.