Memoir

The room where your AI works with everyone.

Memoir is a call with a shared canvas and an agent in the room, and it works for everyone on the call. Your voice is the prompt: say you will pull up the deck and it appears on your screen, for you alone; ask it to draw a system, screen or plan and it appears on the canvas as you talk; ask what was said and it answers from the call itself.

Join the waitlist

In a Memoir room

It works while you talk

Memoir works in real time, during the conversation rather than after it. It listens to everyone who lets it, and what people say is the prompt: no typing, no waiting for a summary. To ask it something, say its name.

Like a just-in-time compiler, it starts with what it can do right away and brings in a better version when that is ready. The first marks land about a second after your words; a slower, more careful pass reads them over and offers a better version beside them only when it finds one. Nothing on the call waits for the slow pass.

The same idea runs on your Mac. When you say you will pull something up, the private assistant starts looking while you are still talking, and the file lands on your screen, for you alone until you press Return.

And like a compiler that drops a guess the moment it proves wrong, Memoir does not defend its mistakes. A written answer from the call shows the lines it came from.

What Memoir never does

Guests see a plain label that Memoir is on and choose whether it hears them.

Join the waitlist

A small group of founders and their teams have Memoir first. Tell us who you are and we will write when a seat opens.