This work proposes an innovative methodology for teaching Quantitative Methods, centered on active learning and using computer programming and Artificial Intelligence (AI) assistants as key technological enablers. Designed for practical courses, this strategy aims to develop skills in computational thinking, algorithm programming, and problem-solving, while at the same time fostering a deep understanding of theoretical concepts.

The methodology is organized into three main phases: Programming Assistance, Adjustment and Validation, and Analysis and Reflection, enhancing practical learning and active interaction with the content.

In the Programming Assistance (PA) phase, students interact with an AI assistant that supports them in designing and developing programming code to solve optimization problems using various algorithms. The AI assistant helps create and explain the programming code using templates provided by the instructor as starting point. Templates are designed to ensure consistent results and help students understand each component of the process, preventing them from simply requesting complete code from the AI. This approach empowers students to tackle programming challenges in a practical and accessible way, ensuring a solid comprehension of each process component while fostering active engagement with the material.

In the Adjustment and Validation (AV) phase, students test and fine-tune their code in an interactive environment, evaluating how modifications affect the obtained results and exploring different representations of the solutions. This stage strengthens practical learning and enhances critical analysis skills by focusing on the impact generated by varying input data and problem characteristics. Focusing on higher added value tasks like fine-tuning and validation, the activity promotes human oversight and an adequate use of AI assistants where the user remains in control.

Finally, the Analysis and Reflection (AR) phase encourages students to discuss their results and delve into advanced topics such as algorithm performance, or the scalability of proposed solutions. This approach fosters the development of critical thinking skills and reinforces their understanding of fundamental concepts through experimentation, results evaluation, and idea exchange.

The implementation of the proposed methodology incorporates various technologies: Python as a base language for its flexibility and ease of use; Jupyter Notebooks provides an online environment for dynamic interaction with code; AI assistants such as ChatGPT facilitate template completion, code analysis, and reflection; and libraries like NumPy, Pandas, and Matplotlib that support data analysis and visualization.

Examples include the implementation of algorithms to solve problems such as Sudoku using a Genetic Algorithm, analyzing different crossover and mutation methods, and the application of Gradient Descent to nonlinear optimization problems, exploring different starting points and learning rates. All learning materials are available as open source resources.

Overall, this methodology fosters active and deep learning, motivating students to apply their knowledge of Quantitative Methods to real-world problems and facilitating the acquisition of key digital skills for their professional futures in e-STEM (entrepreneurship - Science Technology Engineering and Mathematics) profiles.

Keywords: Active Learning, Quantitative Methods, Interactive Learning Environments, AI-Assisted Learning, Problem-Based Learning, Digital Competencies in Higher Education.