📄Small Batch Size Training for Language Models

A simple reference implementation of the single-worker MuLoCo optimizer in Jax & PyTorch. MuLoCo-1 has been shown to outperfrom Muon and have larger critical batch sizes.

Create 3Blue1Brown-style ML research videos with Claude Code. Edit CLAUDE.md, chat with Claude, get publication-quality Manim animations.

AIRS-Bench: an AI Research Science benchmark for quantifying the end-to-end AI research abilities of LLM agents

An efficient implementation of learned optimizers in PyTorch

[TMLR 2025] Meta-learning Optimizers for Communication-Efficient Learning

Simple Learning to Optimize in PyTorch

Tools to connect to and interact with the Mila cluster

A playbook for systematically maximizing the performance of deep learning models.

Hydra is a framework for elegantly configuring complex applications

Code of the CVPR 2021 Oral paper: A Recurrent Vision-and-Language BERT for Navigation

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

JAX - A curated list of resources https://github.com/google/jax

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Code for the paper "Improving Vision-and-Language Navigation with Image-Text Pairs from the Web" (ECCV 2020)

PyTorch extensions for high performance and large scale training.

PyTorch implementation of "The Option Keyboard: Combining Skills in Reinforcement Learning" (NeurIPS 2019)

Official code for paper "Spatially Aware Multimodal Transformers for TextVQA" published at ECCV, 2020.

[CVPR 2021] VirTex: Learning Visual Representations from Textual Annotations

Provides everything needed for high performance data loading and augmentation in pytorch.

Shape and dimension inference (Keras-like) for PyTorch layers and neural networks

Official PyTorch implementation of "GlobalTrack: A Simple and Strong Baseline for Long-term Tracking" @ AAAI2020.

Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

Move and resize windows on macOS with keyboard shortcuts and snap areas

Resources I used for ML Engineer, Applied Scientist and Quant Researcher interviews.

PyTorch Tutorial for Deep Learning Researchers

IKEA Furniture Assembly Environment for Long-Horizon Complex Manipulation Tasks

A flexible, high-performance 3D simulator for Embodied AI research.

Example of using the overcap partition/interruptible jobs