liked that it mirrored a real end-to-end data science workflow — exploring a raw booking dataset, engineering features, training an interpretable model (Random Forest), and then translating the technical results into a single, clear slide for a non-technical stakeholder. It was a good exercise in balancing analytical depth with business communication.

I liked how the simulation used real British Airways customer booking data to solve a practical business problem. Building a Random Forest model to predict booking completion and understanding feature importance like booking_origin and route was very insightful. It gave me hands-on experience in data science and machine learning.

The task used real flight and booking data rather than a purely theoretical exercise, which made it feel close to how BA's Airport Planning and Data Science teams actually work — from building an eligibility model with limited data to training a classification model and presenting results to a non-technical stakeholder.