ChatGPT prompt for creating a data cleaning pipeline for messy CSV files

Act as a senior data engineer who specializes in data quality and ETL pipelines. Help me build a data cleaning pipeline for messy CSV files. Data context: - Source: [where the CSV comes from — CRM export, web scraping, manual entry, etc.] - Approximate rows: [number] - Columns: [list the column names and expected types] - Known issues: [list problems you've noticed — duplicates, missing values, inconsistent formats, encoding issues, etc.] - Target: [where the clean data needs to go — database, dashboard, another system] - Language: [Python (pandas) / R / SQL] Build a pipeline that handles: 1. File loading with encoding detection and error handling 2. Column name standardization (lowercase, underscores, consistent naming) 3. Data type detection and casting (dates, numbers, booleans, categories) 4. Missing value analysis — report missing counts per column, then apply appropriate strategies (drop, fill, interpolate) with justification 5. Duplicate detection — exact and fuzzy duplicates 6. Outlier detection for numeric columns (IQR method + Z-score) 7. String cleaning — trim whitespace, fix encoding, standardize case, normalize special characters 8. Date parsing — handle multiple date formats in the same column 9. Validation rules — define and check business logic constraints 10. Output: clean CSV + data quality report (before/after stats) Code should be modular, well-commented, and include logging. Provide a sample config file so the pipeline can be reused for different CSVs.