My background is in distributed systems and low-level system software. I got my first taste of AI/ML during my master’s from UW-Madison, where I worked at the BadgerRL lab building the early infrastructure to help train humanoid robots learn to play soccer. Since graduating, I have gone back to work on working on building distributed AI infrastructure at Google. However, with the release of ChatGPT and the proliferation of LLMs and GenAI, I have kept in touch with the various developments in the field, and below are the notes I have taken on reading various papers and blogs -