The simplest way for data scientists to greatly improve machine learning performance by enriching their models, both in production or a test environment, with informative external features from any structured or unstructured (text) data source—news, social media, COVID19, economic, financial, geolocation, mobility, and weather data, etc.