Skip to content
View S1ro1's full-sized avatar
67 followers · 32 following

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: StarstruckAchievement: YOLO

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: StarstruckAchievement: YOLO

Highlights

  • Pro

Block or report S1ro1

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
S1ro1/README.md
X Badge LinkedIn Badge

hey

👨‍💻 About Me

Computer science student with passion for ML Systems and AI. Currently in my 2nd year of Master's degree. I work as as a machine learning engineer @ 🤗 Huggingface, where I help to maintain and contribute to Accelerate. When I'm not working, I'm one of the main contributors to KernelBot.

  • 🔭 I'm interested in all things related to AI and ML

  • 🧠 I love to understand things in depth and learn about how the underlying hardware works

  • 🌱 My current focus in on understanding the internals of GPUs and distributed communication

  • 📚 I like to read papers and blogs related to reinforcement learning of LLMs

💻 My Skills

  • 🐍 Been working with Python for more than 10 years and I consider myself very confident in this language

  • 🧠 Familiar with different architectures of neural networks, mainly focusing on causal language modeling

  • ⚙️ I have experience with distributed systems and parallel computing, especially in the context of ML/AI, where I have worked with and in-depth understood different distributed training strategies

  • ⚡️ I love optimizing code and making it run fast, with experience in languages such as C++, Cuda and Triton

Pinned Loading

  1. huggingface/accelerate huggingface/accelerate Public

    🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

    Python 9.2k 1.2k

  2. huggingface/transformers huggingface/transformers Public

    🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

    Python 151k 30.9k

  3. linkedin/Liger-Kernel linkedin/Liger-Kernel Public

    Efficient Triton Kernels for LLM Training

    Python 5.8k 418

  4. gpu-mode/discord-cluster-manager gpu-mode/discord-cluster-manager Public

    Write a fast kernel and run it on Discord. See how you compare against the best!

    Python 58 15

  5. gpu-mode/popcorn-cli gpu-mode/popcorn-cli Public

    Rust 42 13

  6. VideoSuperResolutionNet VideoSuperResolutionNet Public

    Python