Ljómi: 10 Years in the Making

Over a decade ago I saw the Aurora Borealis in Iceland. It was one of those experiences that rewires something in you. Arcs of glowing green ribbons ran horizon to horizon, shifting and pulsing overhead in a way that felt alive.

Standing under it, I started thinking about how ancient peoples would have read the Northern Lights. With no framework to explain what they were seeing, the aurora must have felt godlike, a presence rather than a phenomenon. And I wondered what it would sound like if it could speak.

The idea

Years later, talking it over with a friend, the idea took shape. What if we could convert the aurora into audio? Not a simulation, and not a soundtrack laid over the top, but a real-time translation, with the actual colours, movements and patterns driving the music directly.

I imagined something atmospheric and otherworldly. The Lights producing a sound that felt as vast and shifting as they looked.

The problem was that I didn't have the technical expertise to build it. The idea went on the shelf.

What changed

What made Ljómi possible was AI-assisted development tooling. Cursor, Windsurf, Claude and Deepseek changed what one person could build alone. They didn't write the software for me, but they let me go from concept to working code at a pace I couldn't have managed a few years ago.

After several months of specification work and development, I had a working system. Ljómi, named after the Icelandic word for light or gleam, could take a live aurora feed and produce MIDI output in real time.

Where we are now

Ljómi reads aurora footage for colour (green, pink, purple), intensity, movement direction and speed, wave patterns, and area coverage. It translates all of that into notes, chords and MIDI CC values you can route to any instrument or synth.

Ten years sat between the idea and the thing itself, and none of that gap was the idea's fault. It was always sound. The tools just took that long to catch up.