Maven was tasked with using machine learning to identify vehicles and other objects in drone footage, taking that burden off analysts. Maven’s initial goal was to provide the military with advanced computer vision, enabling the automated detection and identification of objects in as many as 38 categories captured by a drone’s full-motion camera, according to the Pentagon. Maven provides the department with the ability to track individuals as they come and go from different locations.

The Defense Department set an aggressive timeline for Maven—the project was expected to be up and running just six months after it was founded, and reportedly has been deployed in the fight against the Islamic State since December.