This work presents a methodology relying on Artificial Intelligence (AI) and modelling methods to improve the synthesis of new coating solutions for the surface finishing industry. Such methodology is based on the development of a hybrid dataset composed of experimental data of electroplating processes and also computational models. These models describe the hydrodynamics of the system on the one hand, and the coating deposition on the other. Such dataset is used with two main purposes: first, feeding a Generative Model (GAN) for the design and optimization of the electrolytic bath and second, to feed a Physics Informed System (PINN) that predicts properties of the coating deposition for a set of electrochemical conditions. The present work describes the structure of the dataset and the main approaches followed to construct the GAN and the PINN networks for each case. This methodology represents a remarkable step towards the definition of a safe and sustainable by design (SSbD) framework for the development of composite coatings that can replace hard chromium, known as a toxic component, for applications in manufacturing, machinery and the automotive industry.