Feng Xie

Feng Xie is senior director at NVIDIA leading the AI compute team for cutting-edge full-stack computing acceleration technologies for AI, from applications to libraries to hardware architecture. His work includes quantization and model compression, framework optimization, code generation technology, and next-generation GPU hardware feature research.
Avatar photo

Posts by Feng Xie

Developer Tools & Techniques

Achieve CUTLASS C++ Performance with Python APIs Using CuTe DSL

CuTe, a core component of CUTLASS 3.x, provides a unified algebra for describing data layouts and thread mappings, and abstracts complex memory access patterns... 9 MIN READ