Hi, I'm Hussain.

Why Ollama suddenly ran everything on the CPU on my AMD APU box, and the one environment variable that brought the GPU back.

How I integrated LLMs into my PKM converter, refactored API abstractions, debugged network/port quirks, and compared local model performance.

Why and how I built a custom Python migration CLI to extract semantic note structures from Logseq and bridge the gap to document-based PKM vaults.