What the client needed
Two government bodies needed language models and could not send data outside. We deployed the models on their own infrastructure.
Our role: technology provider for the modules, backends, apps and services described here.
What was built
- Model deployment on client servers
- No data leaves the client network
- Delivered under NDA. References on request, with client consent
Related capabilities
How the system fits together
Private LLM · On-premise. The drawing shows the main parts and how work moves between them.
What government projects demand
Procurement rules come first
We scope and document to the tender. References, certificates and the project register are ready for evaluators.
Data stays in the country
Systems and models run on government infrastructure or a national cloud.
Officers use it, not engineers
Field apps and screens are built for the people who do the work, in their language.
How we ran it
Related work
AI that reviews documents for a mobile financial service
A dispatch engine that assigns orders with machine learning
A sales and distribution app with AI route planning and data bots

