Skip to content
View kushalthaman's full-sized avatar
29 followers · 9 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Organizations

@physoly

Block or report kushalthaman

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
kushalthaman/README.md

I'm Kushal. My research focuses on improving the systematic generalization, oversight and understanding of the training processes of large language models.

Previously, I worked on the science of neural network training and on improving the Transformer architecture at Stanford NLP, and worked on RL at Applied Compute and Prime Intellect.

Reach out if you'd like to chat about anything!

Pinned Loading

  1. PrimeIntellect-ai/prime-rl PrimeIntellect-ai/prime-rl Public

    Async RL Training at Scale

    Python 738 122

  2. deltakit deltakit Public

    CLI & library for scalable data infrastructure written with Delta Lake

    Rust 4

  3. polysemanticity polysemanticity Public

    Jupyter Notebook 5

  4. stillwater stillwater Public

    End-to-end pre-training data curation toolkit built on Smallpond (DuckDB, 3FS, Arrow) and Ray

    Python 2

  5. 3fs-csi 3fs-csi Public

    k8s CSI node plugin that exposes 3FS to pods by bind-mounting subdirectories from a single node-wide 3FS FUSE mount

    Go 3

  6. 3fs3 3fs3 Public

    an S3-compatible gateway for 3FS

    Rust 3