What the client needed
A predictive service platform for a telecom operator. It anticipates what a customer will need before they call.
Our role: technology provider for the modules, backends, apps and services described here.
What was built
- Predictive customer service platform
Where the AI works
Related capabilities
How the system fits together
Predictive AI · Telecom. The drawing shows the main parts and how work moves between them.
What telecom projects demand
The billing system is old, and it stays
We integrate with legacy billing instead of replacing it.
Millions of users on low-cost phones
Apps are kept light and tested on low-cost Android devices and slow networks.
Every support call costs money
Self-care, assistants and prediction reduce calls to the contact centre.
How we ran it
Related work
Language models running inside government networks
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

