joshuago’s ai-research Bookmarks

A set of advanced theoretically grounded quantization algorithms that enable massive compression for large language models and vector search engines.

Built on a fully open, end-to-end data pipeline that spans pretraining, post-training, and interactive reinforcement learning, which gives developers reproducible building blocks.

HipKittens is an opinionated collection of programming primitives to help developers realize the hardware's capabilities. Includes optimized register tiles, 8-wave and 4-wave kernel patterns instead of wave-specialization to schedule work within processors. Also includes chiplet-optimized cache reuse patterns to schedule work across processors.