Ask HN: Resources and tips for getting into Machine Learning Research

Hello again HN, I'm interested in getting into machine learning, but purely for the purpose of contributing to the field. Short term I have no plans of moving to a machine learning job or pursuing an academic career, so what I'd like to focus on instead is on building a solid knowledge foundation and eventually the ability to understand and contribute to on-going research.

I'm currently a software engineer with an electrical/computer engineer background. Had a few courses on signals and machine learning but they left a lot to be desired.

Currently I've started by going through the Fast.ai foundations course which seems like a good start.

I'd like some suggestions on the following: 1) Useful resources for the math background needed. I would prefer resources which actually go into the gritty details as I already have some math background, but also appreciate resources with good insight on how to learn about concepts eg. 3B1B.

2) What path would you suggest to get a decent overview and a solid foundation on the field? For example, suppose you were starting in the field with hindsight, what knowledge would you seek as a foundation and how would you get an overview of the field?

Thanks in advance