Best resources, practices, and/or libraries for Elixir + AI/ML

Hello, I hope this is the right place for this question/topic. Basically I’m already in the process of learning Elixir and enjoying it a lot… I’m still far from being an expert, but at least have a good enough grasp that I’ve found some open source repos out there where I can actually follow what’s going on.

This year one of my goals is building something relatively simple but non-trivial that involves AI/ML and Elixir. It doesn’t have to involve LLMs and could be something like a system that calculates credit default risk based on a person’s history (something I’d guess would better fit a more classic ML regression approach rather than bringing an LLM in), but really just about anything sounds good. In general I am familiar with but don’t have any REAL experience with tools such as Spark, Pytorch, Tensorflow, LangChain, etc, and I know that for the real world those tools are what a lot of folks would reach for, but my goal includes focusing on learning in the context of Elixir, so I want to keep things as much as possible in Elixir land…

So my question is if anyone has recommendations on resources, libraries, or any insight that could help getting started?

I’ve found these more than useful:

A Practical Guide to Machine Learning in Elixir

  • Chris Grainger:

Sean Moriarty (I recommend his book also)

Andrés Alejos (& blog)

Thomas Millar

Theres also a fair bit on YT:

Using NeuralProphet, Nx and Livebook to Forecast Business Data in Elixir

  • ElixirConf US 2024 - George Guimarães:

Train a Neural Network in Minutes with Elixir & Axon

Dockyard

Nice! And it’s all recent. Thank you so much :pray:

Also have a look at these threads too:

Good luck and keep us posted on what you end up building :023: