← Field Notes
EN/ES

Your WiFi Is Already Watching the Room

May 16, 2026via github · @ruvnet
AIopen-sourcetoolsautomation

What's happening here

Your WiFi router is constantly sending invisible radio waves through every wall in your building. Normally those waves just carry your internet traffic and nobody pays attention to the shapes they bounce off. RuView pays attention.

This open-source project, built on academic research from Carnegie Mellon University, turns those bouncing radio signals into spatial awareness — a kind of radar you already own. Plug in a tiny sensor (about the size of a matchbox, costs $8–9) and the system can tell you if someone is in the next room, whether they're breathing, even a rough outline of how their body is positioned. No camera. No microphone. Works in total darkness.

What makes this especially interesting right now: you don't need the hardware to try it. The whole thing runs on a regular computer today, in a kind of simulation mode. And it works as a plugin inside Claude, the AI assistant — meaning you can ask it questions about what it's sensing, in plain language.

For a hotel, a care home, a warehouse, a gym — anywhere you want to know if spaces are occupied or if someone needs help — this is a genuinely different way of thinking about it.

The project just crossed 57,000 stars on GitHub, which in that world is a sign that a lot of serious people are paying attention.

Words worth knowing

ESP32 — A tiny, cheap microchip originally designed for WiFi. Think of it as a very small computer the size of your thumbnail. Developers love it because it costs almost nothing and can do surprisingly powerful things.

CSI (Channel State Information) — When WiFi signals travel through a room, they change shape slightly depending on what's in the way. CSI is the technical term for reading those changes — like hearing the difference in how sound echoes in an empty vs. a full room.

Edge AI — AI that runs on a small local device rather than sending everything to a distant server. Faster, cheaper, and more private.

Open-source — The code is freely available for anyone to read, use, or adapt. No license fees, no vendor lock-in.


If you run a space where knowing occupancy or wellbeing matters — a clinic, a hotel corridor, a gym floor — it's worth bookmarking this one. The hardware costs less than a good lunch.

Check it out →

Written by David at AC0.AI. Follow on @ac0hero

Want us to audit your site? Takes 60 seconds →