I have switched back to my mac studio as a streaming/recording machine after tying out omarchy for a while. some notes:
omarchy handled the basics just fine, and I was surprised with how easily I got things set up. But stream deck, prompter, and a couple scene level things
Next stream series: building an LLM from scratch.
A lot of us "know how llms work." Prompts, context windows, tokens, kv caches. But if you're like me, you have a lot of holes in deeply understanding the math, training, and the way inference actually works.
Given how
