Cagil is working on machine learning approaches using the knowledge of physical systems to verify the outputs of the algorithms, repeating the process with additional input until the result match input conditions as well as physical laws governing the sample.