Real-world data is rarely clean. Before building models or drawing conclusions, data scientists spend a significant amount of time handling missing values, correcting inconsistencies, removing duplicates, and fixing corrupted entries. This competition is designed to help participants develop those foundational skills.

In this challenge, you will work with a messy tabular dataset that simulates common real-world data quality issues. Your goal is to clean the data and build a predictive model. Performance on the leaderboard will depend on how effectively you prepare the data and extract a meaningful signal.