You receive a dataset that contains duplicate rows, inconsistent formatting, and several missing values. Walk me through the process and specific tools or techniques you would use to clean and prepare this data for analysis.

I start by profiling the data to understand size and types — for example, a recent 120k-row file showed 12% exact duplicates and an 11.5% missing-rate. I’d use Power Query or Python (pandas) and SQL for repeatable steps. First I remove exact duplicates, then apply fuzzy matching (fuzzywuzzy or dedupe) for near-duplicates, which cut duplicates by another 86% in a past project. Next I standardize formatting: trim whitespace, normalize case, parse dates with strict formats, and convert numeric types. For missing values I set rules: drop rows with >50% missing, impute median for continuous fields, mode for categorical, and always flag imputed values. Finally I validate with constraints, create checksums and a data dictionary, and document the pipeline so it’s reproducible — the whole cleanup took about 3 hours and reduced missing data to under 1%.