Beautifully designed wrapper around Tinker for browsing, probing, and switching LoRA checkpoints across experiments.

SWE-Marathon: an ultra long-horizon SWE benchmark

SIA is a Self Improving AI framework to autonomously improve the performance of any AI system (Model / Agent) on a benchmark task.

One-click Rama cluster deploy on AWS

Autoware - the world's leading open-source software project for autonomous ship

⚙️ Create and run workflows (RPA 2.0)

The fastai book, published as Jupyter Notebooks

Personal portfolio website.

MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

A high-throughput and memory-efficient inference and serving engine for LLMs. Aim: Adding caching.

🐉 Making Rust a first-class language and ecosystem for GPU shaders 🚧

A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

Website that has a location based news search. This means that the news specific to the locality that the user selects will be displayed.

Connecting via wifi to message via the program

Decentralized RL Training at Scale

Medium clone. Want to add more functionality to this.

Radically different form of playing chess. The game is called Angles & Demons.

An aggregation of CTF challenges for csictf.

This projects deals with the containerization of a project.

Post-training with Tinker