Researchers at the University of Toronto have developed a novel end-to-end framework to fast-track the discovery of new structural materials. This framework uses machine learning to accurately predict material properties from their microstructures, and a genetic algorithm to design optimal material microstructures with target mechanical properties. In less than eight hours, the researchers identified titanium alloy microstructures with high strength and stiffness for aerospace components, and compliant microstructures for biomedical implants compatible with human bones.