otternimus
deployedThe AI brain for local work — a home server that keeps private jobs private, with a voice-driven assistant front end (brain + voice) on top. It runs llama.cpp inference over the LAN as the local tier of a three-tier intelligence stack (rule-based logic on the Pi, frontier models in the cloud, otternimus for anything that shouldn’t leave the house), and Frigate detection over the house camera feed. Woken on demand so it only draws power when there’s work.










