In this paper, one presents the method of applying fuzzy logic in the creation of a controller to be used as an intermediary between the operator and the coffee roaster. The purpose of the controller is to provide data regarding the optimal operating conditions of the coffee roaster to obtain a specific type of roast, depending on the user's preference. The concept is intended so that it can be applied with an acceptable degree of accuracy by both a domestic and an industrial roaster. Even though the roasting process has imprecisely defined steps and many variables, the controller user has the ability to define narrower or more comprehensive rule sets, depending on experience. Most professional roasters benefit from colorimeters, densimeters and other modern equipment to obtain the perfect roast, equipment that is difficult to procure being niche tools and having a very high price. The controller presented in this project is an autonomous device intended to be used before the initiation of the roasting process. Depending on the data entered at the inputs, it generates at the outputs the optimal operating parameters of the roaster, so as to obtain a specific type of roasting. The controller uses a generalized approach that does not take into account the type of equipment used for roasting, but only the reactions that take place in the coffee beans during roasting.