As part of the TEC's educational model redesign, engineering students are required to complete the Differential and Integral Calculus course in an intensive 10-week period. This demands that they arrive with strong mathematical skills to successfully tackle the challenges of such a demanding subject. To address this need, a 5-week precalculus course was designed to strengthen fundamental concepts in algebra, geometry, and trigonometry.

To enrich the learning process, an innovative approach was implemented using artificial intelligence. ChatGPT-4 powered by Wolfram was utilized to create a robust bank of parameterized questions. This approach enabled the generation of personalized exercises by adjusting specific parameters, ensuring a variety of problems with difficulty levels tailored to the students' needs. These questions were generated in LaTeX format, facilitating their implementation on platforms like Quizizz for quick and adaptive testing. The incorporation of this technological tool served a dual purpose: to foster study outside the classroom and to provide a more enriching learning experience through frequent practice. The adaptive platform allowed students to access a larger volume of exercises, promoting more autonomous and continuous learning, which reinforced their understanding of key concepts.

To measure the effectiveness of this educational innovation, a methodological study was conducted using a pretest-posttest design that included both implementation and control groups. The pretest results showed similar initial levels between the two groups, ensuring equivalence of conditions at the start of the experiment. After the intervention, students completed a posttest designed to assess their progress in precalculus topics. The data were analyzed using an analysis of variance (ANOVA), which revealed a significant improvement in the performance of students who participated in the educational innovation.

Keywords: Educational Innovation, Higher Education, Adaptive learning.