Tommy

What it does

TOMMY is a Wi-Fi sensing system that detects human presence through walls and obstacles. It converts ESP32 microcontroller boards into motion and presence sensors by analyzing Channel State Information (CSI) from Wi-Fi packets. The system requires at least two ESP32 devices and a server component to establish a sensing zone. All data stays on the user's local network.

Who it is for

Home automation enthusiasts who own or are willing to purchase ESP32 boards. The product integrates with Home Assistant and Matter, making it relevant to people already running those platforms. Users frustrated with conventional motion sensors that create dead zones or cannot detect stationary presence are the intended audience. The system appeals to builders who prefer not to buy proprietary hardware for each room.

Pricing

The site does not show prices.

How it stands out

TOMMY uses Wi-Fi sensing rather than infrared or radar, eliminating the need for line-of-sight detection or dedicated sensors in every room. The approach leverages existing Wi-Fi infrastructure and works with hardware users may already own, such as smart home relays or bulbs running ESPHome. Newer ESP32 variants like the C5 and C6 detect stationary presence, not just motion. The system is self-hosted with no cloud dependency. Setup involves either flashing a dedicated board via USB or adding the TOMMY component to existing ESPHome devices through configuration files.

What a founder should check

First, verify the competitive landscape for presence detection. Established alternatives include infrared motion sensors, radar-based systems, and commercial presence detection platforms. Understand what switching costs and feature gaps might prevent adoption of a software-only solution that requires purchasing or repurposing hardware.

Second, test the practical moat. Wi-Fi sensing is described as a major research topic with nothing production-ready yet, suggesting the technical barrier may be moderate. Investigate whether the approach scales across different home layouts, Wi-Fi configurations, and interference conditions. The requirement that all devices in a zone use the same Wi-Fi band (2.4 or 5 GHz) is a real constraint worth exploring.

Third, assess pricing pressure. Because the system runs on generic, cheap ESP32 boards rather than proprietary hardware, unit economics may be tight. Consider whether customers will prefer low-cost open-source alternatives once the technique matures, and whether Home Assistant and Matter integrations create lock-in or commoditize the offering.