AI-powered Plex recommendations using a fully local LLM (Ollama). Analyzes each user's watch history and generates personalized movie/TV picks — no cloud, no API costs, no data leaving your server. Also includes: Whisper ASR subtitle generation, Ollama-powered subtitle translation, and library maintenance tools. Requires: Ollama container running with a model pulled (recommended: qwen3.5:9b for 12GB VRAM, gemma3:4b for 8GB). Web dashboard at http://[IP]:[PORT:8000]/