Create a business assistant
An agent that answers your team's questions from your data, created self-service
Created self-service
Autonomous Agents
AI Agents
Autonomous AI workers, inside your perimeter
An agent is a prompt, a model and read-only access to your data. Create it self-service, expose it on the interfaces you enable: it works on your data, with your rules, without ever leaving your perimeter.
Specifications
- Definition
- A prompt, a model, your data
- Data access
- Read-only, scoped by you
- Models
- Local vLLM or European APIs
- Governance
- Inherited from the platform
- Creation
- Self-service
- Exposure
- The interfaces you enable
Use cases
Run inside your perimeter
Local models served by the platform (vLLM) or European APIs: inference runs where you decide
No data exposed
Expose under control
Each agent is only reachable from the interfaces you enable, with read-only access to data
Controlled exposure
In action
A business team creates its assistant
You want an assistant that answers your team's questions from your own data
- Create the agent from the console: its name and what it is for
- Choose what it can do and the data sources it may read
- Write its instructions, then pick the model that runs it
- Enable the interfaces: the console chat, and mail if you need it
A working business assistant, created self-service, with no ticket
IT opens AI to business teams
A platform team lets business teams create their agents without losing control of the data
- Every agent authenticates with a read-only service account: the platform refuses to start it if that account goes further
- The scope stays explicit: an agent only reads the data sources it was given
- Choose where inference runs: models served inside your cluster or an external provider
- Find every agent created in the console, with its model, its sources and its interfaces
Business teams create their agents, IT keeps control of the data scope
Key benefits
- Self-service creation: a prompt, a model, a data scope
- Strictly read-only access to your data
- Local models (vLLM) or European APIs, your choice
- Platform governance applies to every agent
Ready to create your first AI agent?
Discover how AI agents work on your data without exposing it.