Using electronic eye to analyze beer is a common type of characterizing styles by hue in brewery’s laboratories, but using only one parameter is not enough to consistently identify the various types of this beverage. Another barrier is to put the optical analysis inside the brewing process, avoiding this laboratory analyses. This study aims to overcome this barriers by creating and electronic eye which can analyze the hopped wort, the brewing wort and the beer at real time, doing the quality control. In this initial phase, the electronic eye was designed, programmed and built from scratch, with cheap elements; then twenty three beers, in a wide range of styles and labels, were analyze by this device, getting the frontal and lateral HSV color space of each sample, storing the data in a database. Then, an artificial neural network (ANN) code was wrote to classify the data and make predictions of style and label, by only using the beer’s image. The six parameters were weighted and, by this point, the frontal hue and both frontal and lateral values were rated, by the deep machine learning code, the most important parameters of this analysis. This process had very precise and accurate responses using validation tests in about twenty five seconds, from getting the beer image, comparing it to the database and the program giving an accurate answer. Although the project seems promising, there is much more to optimize and improve, to make faster and more accurate analyses. This study can be considered a first application of this combined process inside the brewing process, that can be faster, more accurate and cheaper.