docker containers

Turn idle, low-powered Linux servers into private, cheap AI workers using quantized GGUF models and fast CPU runtimes (llama.cpp, whisper.cpp). This guide picks right workloads, installs minimal toolchains, runs LLM/STT locally, exposes a REST server, wires Bash + systemd timers, and caps resources—delivering log summaries, ticket triage, voicemail transcription, and FAQs you can ship this week.