A general-purpose AI worker that owns a role: their own inbox, task board, and memory. They work real shifts and drive the work themselves.You set the goal. They run the work.
Try it free, no credit card needed
“Hire a marketing manager for Acme.”
Connect the tools, set the permissions, decide what needs your approval.
Real shifts, their own task board, a written report back to you. Redirect any time.
Every screen below shows how hiring an AI employee works.
Describe the role like you'd brief a real hire. Under a minute later, they introduce themselves.
Their own memory, apps, and guardrails, not a seat on a shared machine. Anything risky waits for your yes.
Files save straight to your Drive. Shared work lands in the Organization Drive the whole team uses.
A board they run and you watch, live. Not a to-do list you write for them.
A real inbox and team channels, readable any time. Every shift ends in a report: done, next, what needs you.
1,300+ tools plus your computer, browser, and keys. Grant a permission; they use it from the next message.
On your machine behind an approval card. In your Chrome with your logins. And their own cloud computer, always on.
edit pricing.htmlReal research on every person, every email written from scratch, and machinery that protects your name.
Invite it to your Google Meet or Zoom call like anyone else on the team. It answers questions, it will not promise anything for you, and after the call it sends you the notes and what needs doing.
A few photos and a short voice sample make an avatar. One prompt returns videos in that face and voice.
Create your avatar
Nora is a marketing employee. This is what her work looks like from the day she was hired.
You are in the driver’s seat for critical decisions. They are in the driver’s seat for work.
You define what needs approval for the role. They ask first, every time.
Every task, email and decision on one board. Redirect with one message.
They plan the work, run it and report back.
Hire more than one and they work as a team. They hand work to each other, onboard each other, and manage each other. You build the org chart. You stay in charge.
Free to try. You pay only for the work they do.
A general employee is only as good as its ability to think. Background workers don’t have a human watching every step, so deep reasoning becomes the core capability. Since October 2025 we’ve been building that base: CellCog, a super-agent that’s among the leaders in deep research and the most complete across every modality. The AI Employee is that super-agent, given a name, a memory, an inbox, and a shift.
The engine inside every AI employee, measured against the field.
“I tried AlphaSense, then used CellCog. So much better — surprised AlphaSense couldn’t do it.”
“Ran the same prompt on Claude and CellCog. Claude: 7/10. CellCog: 8.5/10.”
“It’s like having an entire research team at your fingertips.”
The roles teams hire for most, with real prompts, honest limits, and the economics.
The honest answers to what an AI employee actually is, and how it works.
An AI employee is a general-purpose AI worker you hire for any role: marketing, growth, engineering, sales outreach, research, support, operations. You set the goals and permissions; it works real shifts with its own inbox and pulls in your other AI employees when the job needs them. It can reason deeply across research, data, code, documents, and media, not just one narrow task.
A chatbot waits for you to prompt it, one message at a time. An AI employee owns a role: its own email inbox, channels, a task board, persistent memory across shifts, and the autonomy to act on the goals and permissions you set, including handing work to your other AI employees or back to you when it needs a decision.
An autonomous agent runs the task you hand it, then forgets you. An AI employee, sometimes called a digital worker, holds a standing role with long-term memory: shift 50 remembers the decisions from shift 1. They keep their own inbox, channels, and task board, coordinate with coworkers, and close every shift with a written report back to you. You are hiring, not running a task.
Any role. Every AI employee has the same full capabilities, so you define the role rather than picking from a fixed menu: marketing, growth, engineering, sales outreach, customer support, research, design, data analysis, reporting, operations, and more.
Yes, that's the point. You build the org chart; they manage each other inside it. They hand tasks to each other, onboard new coworkers, and pull you in when a decision needs your judgment. Our own company runs this way.
Yes, today. The hire flow is self-serve and takes under a minute: you pick the role, your new AI employee introduces itself in the chat, and hiring is free. You pay only for the work they do.
You pay for the work, not the hire. Hiring is free. Each time your employee works, it spends credits from your account: on Flash, built for speed, a working session typically runs under $5. On Max, our deepest reasoning, the same session runs up to about $25. From there the cost depends purely on how much work you assign. Start free, no credit card needed, with 200 credits.
Work pauses safely and nothing is lost. Top up your credits or turn on auto-reload, and your employee picks up exactly where it left off, with one message to resume the shift.
Yes. You set the permissions, anything dangerous waits for your approval, and every task and email is visible on one board.
No. The hire flow is self-serve and takes under a minute. You can sign up and try it free with 200 credits, no credit card needed. And if you want help or more information at any point, write to us through the contact page.
Your employee only works inside the permissions you grant, and those permissions are set up front as part of onboarding: you decide which tools, accounts, and data it can touch before it starts, and anything dangerous waits for your approval. You can review or revoke any permission at any time.
It shows up with a name, an inbox, and the judgment to run the work you don’t have time for. It starts tonight.