Skip to content
View Yashiru's full-sized avatar
🚀
Low level systems engineering
🚀
Low level systems engineering
53 followers · 96 following
  • @VeggiesLabs
  • Geneva - Switzerland
  • 01:54 (UTC +02:00)
  • X @lgebor

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex3Achievement: Public SponsorAchievement: QuickdrawAchievement: YOLOAchievement: Starstruck

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex3Achievement: Public SponsorAchievement: QuickdrawAchievement: YOLOAchievement: Starstruck

Highlights

  • Pro

Organizations

@TacitML @egre-org

Block or report Yashiru

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Yashiru/README.md
  • 👋 Hi, I’m @Yashiru
  • 🔧 ML performance engineer: low-level systems (Rust, C++, CUDA, Triton) for fast and cheap model training and inference
  • ⚡ Working with GPU programming (CUDA / Triton), PyTorch and Burn internals, and LLM inference (vLLM / SGLang)
  • 💞️ I’m looking to collaborate on Rust, low-level performance, or ML inference projects
  • 📫 How to reach me ?


Pinned Loading

  1. tracel-ai/burn tracel-ai/burn Public

    Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

    Rust 15.7k 1k

  2. Guillotine Guillotine Public

    Ultra-fast byte-pair-encoding tokenizer written from scratch in Rust, benchmarked head-to-head against tiktoken and HuggingFace tokenizers.

    Rust 1

  3. evm-rs-emulator evm-rs-emulator Public

    A blazingly fast in-memory Ethereum Virtual Machine (EVM) emulator written in Rust

    Rust 32 4

  4. warpllama warpllama Public

    Fast, low-memory single-GPU Llama inference in Python, built on hand-written Triton & CUDA kernels.

    Python 1