OpenBlender’s proprietary technology is the only one to enable data scientists to efficiently enrich their machine learning models with meaningful external data from any source (variables from news, social media, financial markets, weather, demographics, etc.) to greatly improve performance. OpenBlender automatically profiles, cleans and transforms structured and unstructured (text) data into a common numeric format ready for ML consumption. Then, users blend their datasets with others from any source that overlap in time or location, transparently adding many new variables to their models. Data is pulled into a Python or R dataframe via open-sourced libraries and an API, which also applies to private data from disparate sources from the customer’s data warehouse. Main use cases: Demand and sales forecasting Supply chain management Marketing analytics Risk management Quantitative modeling IoT and location analytics