I asked my AI agent for a private LLM client/server. It built one out of MQTT and 1,000 lines
I have a PC in my home office with an Nvidia GPU and Ollama on it, and I wanted to ask it questions from my phone while I am out. Every guide I found said roughly the same thing: run something in Docker, put a reverse proxy in front of it, forward a port on the router or install a VPN, provision TLS, add an auth layer, and keep it patched. All of that works. It is also more moving parts than I…
A user wanted to use a private AI model on their computer but was frustrated by the complicated setup process. They asked their AI agent to create a simple solution, resulting in a web page, a small service on their PC, and a rented Linux box running an MQTT broker. The setup works using MQTT, a publish/subscribe message bus, where the phone publishes a question and the PC, subscribed to the topic, asks the AI model for an answer.
The answer is then sent back piece by piece. The internet only knows the broker, which simply facilitates the message exchange. The entire system consists of a web UI with 420 lines of code, a PC service with 94 lines of code, and a Python plugin with 536 lines of code, all totaling less than 1,000 lines of code.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
