Skip to content
View Vixel2006's full-sized avatar
11 followers · 8 following

Achievements

Achievement: Pull Sharkx3Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx3Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: Quickdraw

Block or report Vixel2006

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
vixel2006/README.md

Hi

I'm a Software Engineer with a strong focus on Computer Science, AI, and high-performance systems. I enjoy building things from first principles — compilers, deep learning frameworks, CUDA kernels, and experimental architectures. I learn fast, move fast, and love solving complex problems with elegant code.

🔧 What I Work On

  • Systems & Low-Level Engineering: Rust, C, LLVM, compilers, code generation
  • Machine Learning & AI: Research in Multimodal Learning
  • High-Performance Computing: CUDA, GPU kernels, memory layouts, scheduling

🧠 Current Focus

  • Building Plast, a high-performance compute/runtime system with CUDA backends
  • Contributing to Hugging Face to deepen experience in real ML infrastructure
  • Developing a custom compiler with LLVM codegen
  • Experimenting with Multimodal Learning and ways to have modality-agnostic representations of concepts

🚀 Highlights

  • Started studying university-level math and programming at age 12
  • Comfortable with complex abstractions and fast problem-solving
  • Strong intuition for system design and algorithmic thinking
  • Passionate about AI research, performance engineering, and meaningful software

📫 Connect With Me

Pinned Loading

  1. plast plast Public

    Plast isn’t just another PyTorch knockoff. It’s a deep learning engine built from the ground up — native tensors in C, CUDA-powered ops, a Python-to-CUDA compiler in the works, and a focus on fast,…

    C++ 1 1

  2. GRF GRF Public

    Gated Recurrent Fusion (GRF): a compact, scalable multimodal fusion model that achieves competitive performance to MulT with 3× fewer parameters.

    Python

  3. transformers transformers Public

    Forked from huggingface/transformers

    🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

    Python

  4. tinyc tinyc Public

    a tiny compiled programming language inspired by c, made with rust and LLVM.

    Rust

  5. Markly Markly Public

    Markly is a SaaS platform for managing, summarizing, and retrieving knowledge from the internet. Save articles, research papers, and online content, then let Markly organize it into a structured kn…

    TypeScript

  6. BoC BoC Public

    A research about fusing different modalities into modality agnostic concepts for concept based reasoning and generation

    Python