Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 723 Bytes

File metadata and controls

10 lines (8 loc) · 723 Bytes

Evolutionary algorithms for optimised CUDA kernels.

Posts

  • [Kernel Swapping Affects Model Performance]({% post_url 2026-04-28-kernel-swapping-performance %})
  • [Evolving GPU Kernels with LLMs]({% post_url 2026-04-22-kernel-evolution %})
  • [Diversity Is All You Need (To Converge): Why Evolutionary Algorithms Need Diversity Management]({% post_url 2026-04-09-evolution-diversity %})
  • [AST Edits: The Code Editing Format Nobody Uses]({% post_url 2026-04-02-ast-edits %})
  • [Are you using the right baseline for your kernel benchmarks? An analysis of torch.compile modes]({% post_url 2026-04-16-torch-compile-mode-analysis %})
  • [Open-sourcing kernels on the HuggingFace Hub]({% post_url 2026-05-11-hf-kernel-hub %})