skills

Write fast, idiomatic Polars in Python using the lazy API, and answer natural-language questions about data in one shot. Use this skill whenever a task involves loading, querying, transforming, aggregating, joining, or analyzing tabular data (CSV, Parquet, NDJSON, or DataFrames). top-k, trends, period-over-period, distributions, time series, share-of-total, or writing, debugging, or speeding up Polars code, or translating pandas to Polars. Applies even when the user doesn't say "Polars," and is the default for Python data work when no dataframe library is specified. Covers expressions, window functions, group-bys, joins, the streaming engine for larger-than-memory data, and pandas migration. Not for Polars Cloud, On-Prem, distributed, or GPU workloads.