We utilize various AI techniques such as regression, classification, generative models, and large language models to accelerate the exploration of vast alloy composition spaces and map their composition-structure–property relationships. This approach enables the rapid discovery and design of advanced materials including lightweight alloys, metallic glasses, superalloys, and hydrogen materials. 

To experimentally validate the materials designed by AI, we synthesize alloy thin films and bulk alloys using advanced fabrication techniques such as combinatorial sputtering and additive manufacturing. These approaches enable the rapid exploration of composition spaces and the efficient fabrication of materials predicted by AI models. 

We systematically characterize the structural, mechanical, and functional properties of synthesized alloys using high-throughput measurements and advanced characterization techniques. The resulting datasets are used to map composition–structure–property relationships and are continuously fed back into AI models to further improve materials discovery.