Sales team capturing every discovery call
The AI employee joins the call, takes notes, extracts objections, and drafts the next follow-up before the rep leaves the meeting.
Share a Google Meet, Zoom, or Microsoft Teams link and your employee joins the call as a named participant. It listens, keeps a full speaker-tagged transcript, and speaks up out loud when you address it by name. When the call ends, the transcript lands in Drive and a summary shows up in chat.
Share a Google Meet, Zoom, or Microsoft Teams link in chat and ask your employee to attend, and it joins the call as a named participant, with its own name and avatar, not a generic bot labeled "Notetaker". It takes about a minute to spin up, then waits in the meeting's lobby the same way any guest would; someone still has to click Admit before it can hear anything. Once it's in, it stays for the whole call and builds a full transcript tagged by speaker.
By default your employee joins in active mode: it listens the entire time and also speaks up out loud, in its own voice, whenever you or another participant addresses it by name. Ask for silent mode instead and it only listens and transcribes, never speaking during the call. Either way, it only ever answers when addressed. It does not interrupt or narrate the meeting on its own.
This is different from handing it a recording after the fact, which goes through Audio Transcriber instead. The meeting assistant has to be live in the call while it happens; you cannot point it at a meeting that already ended, only ask it to pull up the transcript afterward. What that live presence buys you is a real participant, not just a text file: you can ask it a question mid-call and it answers using everything said in the meeting so far, the same agent, memory, and tools it uses everywhere else.
Meeting Assistant gives your AI employee a seat in the meeting. It joins on request, waits to be let in, then keeps a live, speaker-tagged transcript for the whole call so you are not relying on memory afterward.
It runs on the same agent as the rest of your workforce, same skills, tools, and memory, so it is not a bolt-on notetaker. Google Meet and Microsoft Teams calls transcribe for free using each platform's own captions; Zoom does not hand captions to bots, so those calls transcribe through the meeting provider's built-in speech-to-text instead. You do not configure anything either way, it is automatic based on which platform you're on.
The assistant turns a call into a readable summary with decisions, objections, action items, and next steps already organized. You do not need to clean up raw notes before sharing them with your team.
The moment the call ends, it drops a short recap into the same private chat thread it always uses with you, no separate meetings inbox to check. From there it can draft follow-up emails or recap messages so momentum continues without manual cleanup.
Every meeting's full transcript is saved as a file in your employee's Drive, under Documents > Meetings, so you can open it directly or find it with a normal file search.
Only a short written summary of what was decided folds into the employee's longer-term memory; the full raw transcript does not. That keeps its memory useful for recalling decisions and context without turning every meeting into a wall of unrelated detail.
If nobody admits your employee from the lobby within 10 minutes, it leaves on its own and tells you plainly why, instead of sitting there silently. The same 10-minute limit applies if the meeting's host never starts the call. If everyone in the room goes silent for 10 minutes straight, it leaves rather than staying on an empty line.
If your employee ends up alone in the meeting because everyone else has left, it exits within 30 seconds so it is not burning credits in an empty room. A live meeting bills by the minute the whole time it is on the call; if your credit balance runs out mid-meeting, it ends the call and tells you why in chat rather than continuing to run for free.
Your AI employee joins the meeting, captures the conversation, and turns the call into follow-up work instead of a dead transcript.
Set the meetings you want the assistant to watch and it joins automatically on the schedule you define. During the call it records the conversation, tracks who said what, and keeps a transcript that can be reviewed later. The assistant is not just a recorder. It is a participant with memory and context.
When the meeting ends, it produces a structured recap: what was decided, what needs follow-up, and what should happen next. That recap can be shared immediately, stored in your workspace, or pushed into downstream workflows like tasks and follow-up drafts. The meeting becomes a work artifact, not a dead file.
The AI employee joins the call, takes notes, extracts objections, and drafts the next follow-up before the rep leaves the meeting.
Recurring check-ins get summarized and stored so the whole team has a searchable account history.
The assistant records each interview, creates a readable summary, and makes comparison easier across candidates.
Board meetings and leadership syncs turn into structured follow-up work instead of notes that get lost in inboxes.
| Before | After |
|---|---|
| Meeting notes are written after the call and depend on someone remembering what mattered. | The assistant captures the conversation live and organizes the important parts automatically. |
| Summaries are inconsistent because different people take notes differently. | The AI employee produces a consistent format every time. |
| Follow-up work gets delayed because the recap is written later. | The assistant drafts the recap as soon as the meeting ends. |
| Past meeting context is buried in video files or random docs. | Everything stays in a searchable workspace tied to the meeting. |
It joins the moment you share a live meeting link in chat and ask it to attend. It does not run on a calendar or auto-join meetings you never gave it, and it cannot join a meeting that already ended. Once it's in, it waits in the lobby like any guest until the host admits it.
Yes, all three. Google Meet and Teams calls transcribe using the platform's own free captions; Zoom calls transcribe through the meeting provider's built-in speech-to-text instead, automatically, with nothing for you to configure.
By default it does both: it listens the whole time and speaks up out loud, in its own voice, whenever it's addressed by name. Ask for silent mode if you want it to only listen and transcribe without ever speaking on the call.
It waits in the lobby for up to 10 minutes. If nobody admits it in that time, it leaves and tells you plainly why in chat, rather than sitting there indefinitely.
Yes. Full transcripts are saved as files in your employee's Drive under Documents > Meetings, so you can open or search them directly. Its longer-term memory keeps only the short summary of what was decided, not the full transcript.
Yes, a live meeting bills like any other work your employee does, per minute it is actually in the call, not per meeting. If your credit balance runs out while it is still in a meeting, it leaves and tells you why instead of continuing for free.
It joins under your employee's own name and avatar, not a generic label like "Notetaker". Everyone else in the call sees exactly which of your employees is attending.
We used to spend 20 minutes after every call turning notes into a summary and follow-up email. Now the assistant gives us both before anyone has time to drift to the next task.
Send an Employee to Your Meeting is part of Things they can do out of the box.
Every AI employee ships with built-in abilities that work out of the box. Search the live web for real-time information, scrape and crawl any website for structured data, and generate images directly in conversation. No plugins to install, no APIs to configure.