The new CUTLASS Python DSL and compiler stack is about to be presented at GTC! CUTLASS is our CUDA framework for matrix computations. This is the future of tensorcore and GPU programming. You can stream the talks live or watch the on demand video later. www.nvidia.com/gtc/session-...

We just announced cuTile, a tile programming model for CUDA! It's an array-based paradigm where the MLIR compiler automates mem movement, pipelining & tensor core utilization, making GPU coding easier & more portable. I've been lucky enough to work with a really amazing team that's behind this!