Deep Learning Compiler Engineer
Core
Design and implement compiler transformations, MLIR-based dialects, and lowering passes for CUDA Tile and TensorIR to optimize high-performance GPU kernels.
Role type
Deep Learning Compiler Engineer
Builds
CUDA Tile and TensorIR compiler infrastructure for NVIDIA GPUs
Domain
Deep Learning, GPU Computing, Compiler Infrastructure
Deliverable
production ML models
Required skills
C/C++ programming, compiler optimization, IR design, performance analysis, MLIR, CUDA programming
Preferred skills
GPU architecture knowledge, OpenCL, LLVM, XLA, TVM
Technologies
CUDA, TensorIR, CUDA Tile, MLIR, LLVM, XLA, TVM
Responsibilities
Design and implement compiler transformations, develop MLIR-based dialects and lowering passes, optimize tile-based kernel performance, define public APIs, craft compiler and optimization techniques
Seniority
Mid-Senior, hands-on IC
