In the digital era of rapid artificial intelligence development, there is an urgent demand in the data mining field for talents with integrated “algorithm + business” capabilities. Addressing issues such as high thresholds for code implementation, shallow understanding of algorithm principles, and difficulties in implementing optimization strategies in traditional data mining teaching, this study focuses on cultivating the data mining capabilities of students in computer majors. Taking the Python ecosystem as the core teaching tool, generative AI tools are organically integrated as auxiliary means into the entire teaching process. Relying on the “Enterprise Customer Churn Prediction” project, the paper designs specific intervention points of AI tools in key links such as data exploration, model construction, algorithm optimization, and result interpretation. Practice shows that this model can effectively reduce programming cognitive load, stimulate students to explore algorithm optimization logic, and improve the quality of model implementation, thereby addressing students’ fear of difficulties, enhancing their comprehensive capabilities to solve complex engineering problems, and providing a practical and referable path for digital teaching reform in computer courses in similar institutions.