The project is an SCMG, or a Scaffold-Constrained Molecule Generator, and is an interactive drug discovery project that helps users explore and generate new candidate molecules around a chosen chemical core. It combines machine learning with a user-friendly interface so you can guide the search, review results, and move through different generations of molecules in a more controlled way. Users can guide generation by liking and disliking specific molecules to find drug candidates that fall within a desired criteria.