Geta.Team Geta.Team

Key concepts

Employee, workspace, dashboard, credits, model tiers, channels, skills. The vocabulary you need.

Seven words carry the whole product. Learn them once and the rest of this documentation reads itself.

AI employee

A worker with a name, a face, a position, and a personality. It has its own email address and phone number, a memory that outlives the conversation, and the skills you grant it.

You create one in two ways: from scratch with New Employee, or from the Employee Catalog, a library of ready-made profiles you can filter by job type, Sales, Admin, HR, DevOps, SEO, Legal, Design, Finance, Video, and more. If none fits, the catalog offers Create your custom employee.

The employee catalog, filtered by job type

An employee's own screen

Workspace

One deployed instance of Geta.Team, on its own dedicated server, in the EU or US region you choose when you create it. It holds its own employees, connectors, settings, and history.

Use several workspaces to separate clients, teams, or environments.

On SaaS Cloud the number is capped by your plan: 3 on Entry, 5 on Solo, 10 on Starter, 20 on the Pro tiers. Reaching the cap doesn't break anything, creating the next workspace is refused until you delete one or move up a plan.

On BYOK there is no cap. Run as many workspaces as your licence covers.

Dashboard

The home screen of a workspace: the grid of employee cards. Every card shows the employee's status and how long it has been idle, its tags, the channels it is connected to, and four actions, Chat, mail, call, and a menu.

Everything else in the workspace hangs off the dashboard sidebar, hiring, Websites, Kanban, Team Workspace, and Settings.

Credits

SaaS Cloud only. Credits are what your employees spend when they think and act. Every SaaS Cloud plan ships with a fixed monthly volume, from 20,000 to 1,350,000.

They are drawn down as work happens, and the gauge at the bottom of an employee's sidebar shows what is left. When the month's volume runs out, employees stop until the next cycle or an upgrade. There is no surprise overage, you are never billed for more than your plan.

On BYOK there are no credits at all. You bring your own provider keys, so you pay Anthropic, OpenAI or whoever you use directly, and we impose no usage cap. Nothing stops at the end of the month, and there is no credit gauge to watch. What you pay us is the plan itself, priced on the number of employees rather than on volume.

Model tiers

SaaS Cloud only. Every employee runs on one of four tiers, and you pick it per employee:

TierBest for
MiniHigh-volume simple work: classification, routing, tagging
StandardRoutine work: drafts, replies, summaries. The everyday default
MediumLong write-ups, nuanced replies, multi-step plans
MaxCreative work, thorny analysis, high-stakes content

Higher tiers spend credits faster. The point of having four is to put your highest-volume employees on Mini and keep Max for the ones doing your hardest work.

Set it in Configure ‹employee› → Intelligence.

On BYOK the four tiers do not exist. The same screen lists the engines and providers you declared under Settings → LLM Providers, and you pick one of those per employee. You control the cost by choosing the model, not by choosing a tier.

The intelligence level selector

Channels

How an employee reaches people, and how people reach it.

Every employee gets an email address and can take phone calls out of the box. The rest you connect once for the whole workspace, in Settings → Connectors: Slack, Microsoft Teams, Google Workspace, Office 365, WhatsApp, Telegram, and more.

The connectors list

Skills

What an employee is allowed to do, beyond talking: browse the web, run code, generate images, publish a page, query a database. You grant them per employee in Configure ‹employee› → Skills, from a catalog or from skills you write yourself.

The skills catalog

Editions

Two ways to run all of the above: we host it, or you do, with your own LLM keys. See Editions & hosting.

On this page