Lecture 40: CUDA Docs for Humans @GPUMODE

Lecture 40: CUDA Docs for Humans @GPUMODE

Uploaded December 2024 | Updated September 2026, 9 minutes ago

https://x.com/charles_irl/status/1867306225706447023
docs.google.com/presentation/d/15lTG6aqf72Hyk5_lqH7iSrc8aP1ElEYxCxch-tD37PE/edit

Lecture 40: CUDA Docs for Humans

Lecture 87: Low Latency Communication Kernels with NVSHMEM

Outperforming cuBLAS on NVFP4

Lecture 99: Distributed ML on consumer devices

Lecture 79 Mirage (MPK): Compiling LLMs into Mega Kernels

Lecture 107: PithTrain

Lecture 58: Disaggregated LLM Inference

Lecture 59: FastVideo

Lecture 93: Cornserve Easy, Fast and Scalable Multimodal AI

Lecture 84: Numerics and AI

Live - Disaggregated LLM Inference: Past, Present and Future

CuTe

Lecture 40: CUDA Docs for Humans