Ibis Portable Analytics

Write analytics once and run it on any backend with Ibis. A pandas-like Python API that compiles to SQL for DuckDB, Postgres, BigQuery, Snowflake, Spark, and more — no rewriting queries per database.

This skill makes Claude an Ibis expert. It writes portable dataframe expressions that compile to native SQL across 20+ backends, so the same analytics code runs on DuckDB locally and Snowflake in production. Covers deferred expressions, joins and window functions, UDFs, backend switching, and interop with pandas, Polars, and PyArrow for hybrid workflows.

When to use

Use when writing backend-agnostic analytics in Python, moving pandas logic to a warehouse without rewriting SQL, or building queries that run on DuckDB and BigQuery alike.

Examples

Portable aggregation

Run on any backend

Write an Ibis expression for monthly revenue by region that runs on both DuckDB and BigQuery unchanged

Switch backends

Dev to prod

Take my Ibis pipeline running on local DuckDB and point it at Snowflake without changing the logic

Window functions

Ranked analytics

Use Ibis to compute a 7-day rolling average of signups per cohort with window functions