Exorust

Hi everyone, I was wondering what books/courses/projects one might do to learn PyTorch.

(I asked this because I saw this thread about CuDA and I wanted the same details here: https://news.ycombinator.com/item?id=44216123)

I'm sure there is wisdom to be learned from people who've successfully executed a technical spike and figure things out? How do I come out on top?

I've been a Software engineer for 5 years now, and recently started LLM dev.

A lot of the huggingface, accelerate and bitsandbytes are exceptionally brittle and newer models somehow get stuck in dependecy hell.

Hi everyone who's running on prem LLMs, what kind of costs are you folks seeing?

I want to learn good stuff in MLOps, but I have no idea where to start?

What kind of Rigs are people building to finetune or host or run LLMs?