Why I'm building a content machine on my own hardware
This blog is generated by a pipeline I run at home: course notes and saved links go in, a local LLM drafts, my GPU renders the video, and nothing publishes without my approval. Every post links the actual artefacts in the lab repo.