Faderbolt
A personal project: turning a DJ mix from something you can only hear into something you can study and edit.
An intermediate DJ can hear that their transitions feel flatter than the ones they admire, but has no way to actually see why. No score, no source material, nothing to study once the mix is over.
The core bet is visual: a 3-band color waveform, bass, mid and high as red, green and blue, the same visual language Serato and rekordbox already trained a generation of DJs to read. That makes an EQ move visible in the audio itself instead of hidden in an abstract automation lane. I'm treating how the gesture data gets captured as a design decision in its own right, since it has to support seeing a mix, editing it without destroying anything, and comparing two DJs' takes on the same moment.
Right now I'm validating the idea against real audio constraints before committing to a full build. Timestretch quality, how good the renders sound, whether people can tell the output apart from real hardware in a blind test. Design and "can we actually build this" are happening at the same time, not one after the other.
Pre-launch, still in research and technical validation. Included here to show range beyond healthcare. Same design thinking, completely different craft.