Content filtering with ML

We enhanced cyan Security’s child protection and content filtering products by incorporating deep learning models.

Our work involved training deep learning computer vision models to accurately classify web page content based on visual features, improving the ability to filter inappropriate content and control access.

Our custom end-to-end machine learning platform managed the entire lifecycle of all machine learning models, from data handling to model deployment. This platform featured interactive data management, automated model training, comprehensive versioning, robust model governance protocols, and a centralized hub for monitoring model performance, tracking versions, and managing deployments.

The Cyan Security suite was utilized by major telecom providers like Magenta to safeguard their business and private customers.