CUTLASS backend for Inductor: PyTorch Compiler Series

CUTLASS is an NVIDIA-provided library of highly-performant GEMM primitives. In this talk, we discuss key concepts behind CUTLASS and demonstrate how to utilize torch.compile to customize these primitives for a specific workload via autotuning and epilogue fusion. Speaker: Michael Lazos Michael Lazos is a software engineer at Meta where he contributes to torch.compile. His expertise spans both graph extraction with TorchDynamo and generating optimized kernels with the backend compiler TorchInductor. Previously, he was at Microsoft contributing to project Brainwave, an FPGA-based inference accelerator. PyTorch Compiler Series In this video series, watch the PyTorch Compiler team share tips and tricks that help you get the max out of torch.compile, torch.export, and related technologies, while enjoying a glimpse into all the cool engineering work that goes on behind the scenes.