Nvidia Senior Deep Learning Compiler Engineer - CUDA 

United States, Texas 

835552748

What you'll be doing:

• Design and implement Pythonic language interface for tile-aware GPU programming

• Optimizing compiler pipelines for efficient execution

• Integrate with AI/ML frameworks

• Develop performance critical primitives for tensor operations and memory operations

• Collaborate with hardware teams to co-design compiler optimizations for emerging GPU architectures, including Tensor Core utilization and distributed execution

What we need to see:

• Bachelor's degree in Computer Science, Electrical Engineering, or related field (or equivalent experience); MS or PhD are preferred

• 5+ years (academic/ industry) experience with ML/DL systems development preferable for compilers

• Strong Python and C/C++ programming skills

• Expert experience in developing or using deep learning frameworks (e.g. PyTorch, JAX, Triton, etc.)

• Strong sense of ownership, fast learner, passion for quality and user experience

Ways To Stand Out From The Crowd:

• Proficiency in Python and C++ for DSL development and low-level compiler internals

• Experience with compiler frameworks (Triton, LLVM, MLIR,TVM) and IR design for GPUs

• Deep understanding of GPU architectures (CUDA cores, Tensor Cores, memory hierarchies) and tile-based execution models

• Strong experience machine learning systems research and productionization

• Open source project ownership or contributions