Optimizing a WebGPU Matmul Kernel for 1 TFLOP(zanussbaum.substack.com)163 points by zanussbaum 1 year ago | 68 comments
Batmobile: 10-20x Faster CUDA Kernels for Equivariant Graph Neural Networks(elliotarledge.com)83 points by ipnon 9 months ago | 11 comments
Gemlite: Towards Building Custom Low-Bit Fused CUDA Kernels(mobiusml.github.io)40 points by un_ess 2 years ago | 2 comments
Were RNNs all we needed? A GPU programming perspective(dhruvmsheth.github.io)88 points by omegablues 1 year ago | 24 comments
How to use Claude Code subagents to parallelize development(zachwills.net)267 points by zachwills 1 year ago | 117 comments
Ultra efficient vector extension for SQLite(marcobambini.substack.com)139 points by marcobambini 1 year ago | 47 comments