Logan Ward University of Chicago Materials Science and Engineering One factor limiting the ability of engineers to develop better materials is the speed at which they can search through possible formulations and processing schemes for materials. Recently, machine learning algorithms have emerged as an effective route for accelerating the search for new materials by suggesting particularly valuable experiments. In this presentation, we will discuss work towards addressing two major challenges in the use of machine learning in materials engineering: (i) the lack of high quality materials data, (ii) the need for general-purpose approaches for using machine learning with materials data. We will first describe a tool for automatically solving crystal structures using the First-Principles Assisted Structure Solution (FPASS) method: a technique based on an evolutionary algorithm. We will then introduce several strategies for using materials data to training machine learning models and demonstrate how machine learning can be applied to optimize existing and discover new Bulk Metallic Glass alloys, among other types of materials.