Katja Sirazitdinova

Katja Sirazitdinova is an NVIDIA AI and accelerated-computing practitioner and speaker focused on generative AI, deep learning, JAX, CuTe DSL, and custom GPU kernels. NVIDIA has profiled her as a senior deep learning data scientist helping customers design, optimize, and deploy AI systems.

CuTe DSL Brings Custom NVIDIA GPU Kernels Into JAX

Katja Sirazitdinova presents CuTe DSL for JAX as an escape hatch for developers who need more GPU-kernel control than XLA provides, without leaving the JAX workflow. The tutorial argues that custom NVIDIA kernels can be written in Python with CUTLASS CuTe DSL, bridged into compiled JAX programs through `cutlass_call`, and used where fusion, special layouts, custom data movement, or narrow performance bottlenecks justify the extra discipline around shapes, dtypes, launch constraints, and static parameters.

NVIDIA DeveloperMay 7, 20267 min read