ICE is an independent client-server platform for content analysis and content engineering. The system integrates different component, such as text-classification, clustering, language identification, duplicate recognition, information extraction, etc. Each component can be configured and parametrised in individual workflows. New modules can be easily integrated and adapted to the workflow The platform includes interactive tools for the development and management of multi-lingual knowledge models, such as hierarchical classification trees, training and test data for AI-based categorisation, as well as rules and parameters for the configuration of components. Knowledge models can be deployed independently and simultaneously in different workflows (such as test or productive environments) or different instances. All functions can be monitored and evaluated through benchmarking devices. The system comes with elaborate reporting mechanisms which allow to view the results of the processing in detailed statistics.