Worked on the pymc-marketing repository to enhance input validation for the CLVModel, focusing on improving data onboarding and reliability. Addressed a critical bug by implementing comprehensive missing-columns reporting, allowing the system to identify all required-but-missing fields and raise a single, descriptive ValueError for incomplete datasets. This approach streamlined error handling and reduced downstream data-cleaning efforts. Leveraged Python, Pandas, and data validation techniques to refine checks for unique and homogeneous column constraints, ensuring higher data integrity. The update enabled more accurate customer lifetime value modeling and provided users with clearer, actionable feedback during data ingestion, supporting faster issue resolution.