so it seems! Great tour and defo would like to hear more about the stack that you run.
Particularly interested in running local AI, and at least getting the initial setup "right". Was thinking about how to use Hermes with a local LLM running on my gaming PC which has a nice powerful RTX5090. LM Studio seems good, I'm not smart enough to configure Ollama properly. Need one of your fab videos!!! 😎