Private AI assistants
Assistants built on open-weight LLMs, deployed on your servers or a private cloud. The model, data, and logs stay in your hands.
A good fit when: documents must not leave the organization.
You know where your data goes
Before we build anything, location and access are agreed in writing for each project. There are only three options, and you choose.
Documents and data you chooseOur default: A or B with an open-weight AI model. C is a documented exception, not a shortcut.
- A
Your organization’s own serversOn-premise. Access is held by people you appoint. - B
A private cloud of your choiceA data center in Indonesia by default. Access stays in your hands. - C
External servicesOnly with your written permission, added to a per-project allowlist. Without permission: not connected.