Zach Nussbaum personal blog

Optimizing a WebGPU Matmul Kernel for 1 TFLOP

zanussbaum.substack.com published 2024-11-11

Building Surfgrad, a high-performant, WebGPU-powered autograd library

Points

172

HN comments

80

Submissions

1

Cited by

0 commenters

Systems & PerformanceWeb & Frontend2024

Related posts from the same era

The secret inside One Million Checkboxes

Nolen Royalty20242,365 points218 comments details

macOS command-line tools you might not know about

Ssaurabhs.org20232,039 points442 comments details

The Cloud Computer

Oxide Blog20231,880 points956 commentscited by 8 details

The Windows Subsystem for Linux is now open source

Windows Blog20251,613 points1,194 comments details

Cloudflare outage on November 18, 2025 post mortem

Cloudflare Blog20251,465 points916 commentscited by 13 details

LLaMA now goes faster on CPUs

JJustine Tunney20241,372 points451 commentscited by 11 details