Infront of with the difficulties faced in making a new drug available to the population, it is essential to seek ways to simplify the process. In silico methodologies are alternatives to benchtop experiments, being frequently used due to their speed and low cost. The present study aimed to formulate a theoretical-practical activity in the Pharmaceutical Chemistry course, where students applied their knowledge of structural modeling and molecular docking to propose bioactive compounds against molecular targets of the SARS CoV-2 virus. The class was divided, and each group presented a drug candidate, the precursors being natural molecules. In total, seven derivatives were designed and tested against different macromolecules, and then an in silico prediction of their physicochemical characteristics was performed. The docking results were positive for all derivatives, in terms of binding energy, mainly GEND with -9.0 kcal mol-1. In addition, the prototypes exhibited good interactions with the amino acids of the respective targets, mainly KAED, QUED and GEND, in addition to presenting adequate physicochemical properties for meeting the Lipinski restrictions. Therefore, this study presented at least three potential inhibitors of SARS-CoV-2, showing the importance of using computational tools in drug design and development, as well as in teaching practice.