Cleaned and prepared a real-world e-commerce dataset (541,909 transaction records from the UCI Online Retail dataset) for analysis using Python and pandas. Removed exact duplicate rows, dropped records with missing Customer IDs, filtered out cancelled orders and invalid quantity/price values, corrected data types (dates, IDs), and standardized text fields like country and product description. The cleaning process reduced the dataset from 541,909 to 392,692 valid rows. Also engineered a new Total Price feature (Quantity × Unit Price) to prepare the data for downstream sales analysis and reporting. Full code, sample cleaned data, and documentation are available on GitHub.