11-09, 11:45–12:30 (America/New_York), Central Park West (6th floor)
Ibis is a pure Python library that lets you write Python to build up expressions that can be executed on a wide array of backends (sqlite, duckdb, postgres, spark, clickhouse, bigquery, and more!). It offers a dataframe-like interface and is more concise and composable than SQLAlchemy when writing interactive analytics code.
We love to use Python in our day jobs, but that enterprise database you have to run your ETL job against may have other ideas. SQL is powerful and ubiquitous, but wouldn’t it be nice if you had all the power of Python AND could also interact with some highly optimized database engines?
Ibis is a pure Python library that lets you write Python to build up expressions that can be executed on a wide array of backends (sqlite, duckdb, postgres, spark, clickhouse, bigquery, and more!). It offers a dataframe-like interface and is more concise and composable than SQLAlchemy when writing interactive analytics code. And it is NOT a templating library, so no injection shenanigans.
If you:
- have had to translate a proof-of-concept from Pandas to PySpark to run on the “real data”
- download a huge parquet file because the upstream data is the result of 500 lines of dense SQL and you’re afraid to mess with it
- are a data-engineer, data-scientist, data-hobbyist, or data-anything
then come and join us for a tour of what Ibis can do for you!
No previous knowledge expected
Gil Forsyth is a software engineer at Voltron Data. He followed the common career path of Japanese language specialist -> administrative assistant -> mechanical engineer -> computational fluid dynamicist -> data scientist -> software engineer -> machine learning engineer -> software engineer.
Gil contributes to several projects in the PyData ecosystem and is a core maintainer of xonsh and helps maintain Ibis. He served as the program chair for the Scientific Computing with Python (SciPy) conference from 2016 to 2020.