A methodology for evaluating a Digital Signal Processing course in the modern Artificial Intelligence era is presented. The course is part of the Bachelor’s program in Artificial Intelligence at the Federal University of Goiás, Brazil. The evaluation process contemplated a design-implement experience based on the CDIO framework for a course module whose contents evolve quickly due to a fast moving AI industry. The evaluation process was designed in cooperation with the Center of Excellence in Artificial Intelligence in Brazil. The course contents consisted of regular lectures for eleven weeks on signal processing topics, such as signal analysis, spectral analysis and synthesis and image processing. The last five weeks were dedicated to the development of a project involving state-of-the-art Artificial Intelligence methods. The learning tools provided to the students included a classroom platform to upload and share materials and a version control system to keep track of project deliverables, such as slides, project documentation and a kanban to measure teamwork. Forty-four students participated in the learning activity. Two surveys were applied to the students at the course completion in order to assess the level of achievement of the learning outcomes. As a result, the students were able to successfully conduct a project-based learning experience addressing challenging AI applications that expanded the limits of the traditional course contents in a systematic way. We designed a learning environment that allowed us to assess hard and soft skills during the learning process, which are abilities demanded by modern industries. Finally, reflections and observations and student feedback were collected to further improve the course module in further iterations.
