Discovering and designing new materials with tailored performance characteristics, which may be amenable to manufacturing at scale, and which can be implemented in technology much faster than the current state-of-the-art, is crucial to addressing a broad range of economic and societal needs. At SSRL, we use a combination of high throughput x-ray diffraction and scattering structural characterization and operando studies of materials’ behavior in order to inform data driven machine learning models, which accelerate our ability to discover new materials with targeted properties and optimize their performance in systems and devices. Our high-throughput and robotic sample handling capabilities allow us to measure orders of magnitude more materials than was possible even five years ago. We have developed machine learning models to automate data interpretation to enable real time data analysis, and active learning algorithms to automate design of experiment for both static and in-situ and operando measurements. These capabilities are enabling a new type of user experiment which uses x-ray measurements as a real time feedback tool to accelerate the discovery of materials with targeted properties.