The research will focus on developing an ab initio molecular dynamics (AIMD) and hybrid reverse Monte Carlo (HRMC) simulation algorithm, augmented by ab initio based energy constraints, that couples with experimental input and feedback, using a series of thin-film amorphous preceramic polymers—a-BC:H, a-SiBCN:H, and a-SiCO:H—as suitably complex and technologically relevant case studies. The unique utility of modern solid-state nuclear magnetic resonance techniques to obtain specific bonding and connectivity information and the sensitive medium-range order information available from fluctuation electron microscopy—a specialized technique based on transmission electron microscopy—will be combined with neutron diffraction and more routine physical and electronic structure characterization methods to provide input and constraints for the simulations. The HRMC modeling efforts will be optimized via particle swarm optimization and subsequently used to train an artificial neural network (ANN) that will predictively link the parameters used to simulate a desired material with the growth parameters needed to fabricate said material. Consequently, the investigators expect to substantially advance the state of the art and surmount traditional challenges associated with (1) identifying non-global potential energy minima for materials produced under non-thermodynamic conditions and (2) aligning simulation and growth process timescales. This effort will benefit technology and society by advancing the science of design of complex disordered solids. The novelty of the proposed effort lies in developing the algorithms and rule-sets that will tie together growth, characterization, and simulation, as well as in developing strategies for mapping (not necessarily reproducing) fabrication conditions and desired properties, and it is this that takes the proposed effort from evolutionary to potentially revolutionary. The PIs also plan to release the AMD program as open source and build a user community around it by ensuring that interested researchers are able to contribute to the AMD codebase. This will allow a wider growth of the project. This aspect is of special interest to the software cluster in the Office of Advanced Cyberinfrastructure, which has provided co-funding for this award.