Roadmap

Built in the open enough that you can hold us to it. Early-access waves start with dictation, because it already works.

Working today

Speak instead of type

Double-tap Ctrl dictation into any app, powered by local Whisper transcription. Plus the capture engine underneath it all: screen and audio recording, screenshots, region capture - built and tested.

Building now

The daily driver

Making voice-first feel inevitable: faster and more accurate dictation, a personal dictionary for your names and terms, and the polish that makes you stop reaching for the keyboard. This is the phase early-access waves ship from.

Next

You ask, it does

Voice control of the machine: open apps, bring windows forward, start the email and put your words in it. Shipped step by step - each step announced here only when it works.

Planned

The memory

The rolling 24-hour screen memory: timeline scrubber, mark in and out, clip export, and - with ambient audio enabled - searchable transcripts of your day. Forgets by default; keeping is your explicit act.

Planned

The AI bridge and beyond

Hand any moment - screen and words together - straight to your AI instead of typing context by hand. Meeting-aware auto-pause. And phones: the same quiet helper following you off the desktop.

Standing commitments

These are not roadmap items. They hold for every version, starting with the first:

Founding users steer this list

The waitlist question - "what do you wish you could do by voice?" - decides what gets built first.

Join the waitlist