Deep Learning Performance Software Engineer
Core
Develop highly optimized deep learning kernels and end-to-end performance optimization using a tile-based GPU programming model.
Role type
Deep Learning Performance Software Engineer
Builds
GPU-accelerated deep learning software (TileGym, Triton CUDA TileIR backend, CUDA Tile)
Domain
Hardware acceleration / Deep Learning / GPU Computing
Deliverable
production ML models
Required skills
C/C++ programming, software design, performance modeling, profiling, debug, code optimization, architectural knowledge of CPU and GPU, tile-based GPU programming
Preferred skills
Python, MLIR, AI agent experience, CUDA or OpenCL
Technologies
CUDA, OpenCL, MLIR, Triton, TileGym
Responsibilities
Develop TileGym, Triton CUDA TileIR backend and CUDA Tile; Develop highly optimized deep learning kernels through tile-based GPU programming model; End-to-end performance optimization through tile-based GPU programming model; Do performance optimization, analysis, and tuning
Seniority
Mid-level (3+ years experience)
