A Modular, Human-Centric LED System That Reacts to Circadian Patterns
The project started because I noticed how much artificial lighting affects focus, sleep quality, and even mood—yet most consumer lights only offer crude presets like “warm / cool / reading mode.” I wanted something smarter, more adaptive, and more open.
Key ideas behind the project
Circadian-aware light engine: A small local model predicts the ideal light temperature and intensity throughout the day, not just based on time but on actual behavior patterns.
Modular physical design: Each light module has its own driver + sensor bundle (ambient light, motion, color, noise levels). They magnetically attach and can sync or operate independently.
Local-first control: No cloud reliance. Everything runs on-device via a low-power microcontroller and a tiny inference model.
Contextual lighting modes: The system learns to differentiate between “late-night focus,” “winding down,” “creative work,” and “ambient mood,” then shifts lighting accordingly.
Open API: You can script scenes or behaviors using a simple JSON-based API. (Example: “If desk occupancy > 20 min and sound < 40dB, switch to 4200K focus mode.”)
Why develop this?
Most smart lights feel like toys or UI-driven gadgets. I wanted something that behaves more like a quiet assistant—ambient automation rather than app micromanagement.
Also, lighting tech has tons of unexplored potential. With LEDs as cheap and programmable as they are, it feels like an interesting frontier for personal well-being and workspace design.
What I’m looking for
Feedback on the hardware architecture and whether the modular approach makes sense
Suggestions on open-source licensing or sustainability for a small hardware/software hybrid project
Thoughts on whether people would actually use something like this—especially developers, creators, or remote workers
Anyone who has worked in lighting science, IoT hardware, or color perception research—I’d love your insights