The application of machine learning to an extensive, continually expanding dataset from Embry-Riddle Aeronautical University’s flight training program shows significant promise for enhancing the quality and effectiveness of instruction—measured by reduced time to achieve tasks—and potentially lowering costs for both students and the institution. This paper outlines the end-to-end workflow, including flight operations, data acquisition, storage, data preparation, analysis, and reporting. It provides details on the data analysis methods employed, the types of questions addressed through the data, and the intended users of the resulting reports. Preliminary findings related to fleet maintenance and targeted learning objectives are discussed. The paper also examines the challenges encountered during the development of the Prescott Synchronized Training and Learning Environment (PSTLE) and presents some of the solutions implemented. Notably, the project will now incorporate data from simulators and voice transcriptions from Air Traffic Control (ATC), which are synchronized with flight data to enrich analysis. Finally, the paper highlights future directions and poses key questions that must be addressed to transition from prototype to production, inviting collaboration with researchers in academia as well as partners in the public and private sectors.