Nonlinear recording.
Nonlinear editing freed the edit from the timeline. Nonlinear recording frees the take: flub a line, just say it again, and the bad take never existed. The recording you deliver is not the timeline you performed.
Here's the fastest way to share a file from your Mac.
Drag it onto the pill and the link lands on your clipboard. — onto the dock—?
[ go silent — drag the file onto the pill ]
Paste it — and that's exactly what they see.
Moment by moment.
Three decisions you'd normally make in an editor, made instead by talking — while the recorder is rolling.
Words dim as you speak them, like a karaoke cursor.
The script follows your voice
Write a script, press record, read. On-device speech recognition drives the cursor — no scrolling, no pedal, no pace to keep.
Say the line again
Flub and re-speak — the recorder rewinds the cursor and cuts the bad take out of the recording, live. The timer jumps backward: it shows the length of your video, not your session.
Silence never ships
Scripts can hold [stage directions]. Go silent, do the thing on screen, start talking — the dead air is cut the moment you resume, seam landed in a breath.
THE GATE — per-word timestamps tell deliberate re-speaking apart from the mic re-hearing recent audio. That gate is the patent-pending core.
THE SEAMS — cuts anchor to word onsets measured live, then snap into real breath gaps. Every seam lands in silence.
THE PRIVACY — a local speech model, on-device, always. Nothing you say leaves your Mac.
There is no afterward.
Every screen recorder makes you choose: ship the flubs, or fix them afterward. Nonlinear recording removes the afterward.
One take, straight to a link. Fast — and every flub ships with it.
rough by designRecord raw, then fix it afterward in an editor. The take is raw material.
the afterward is the productRetakes and dead air were resolved while you performed. Stop, and the video exists.
there is no afterward