Cortex AISQL: A Production SQL Engine for Unstructured Data Conference: ; ;

Paweł Liskowski, Benjamin Han, Paritosh Aggarwal, Bowei Chen, Boxin Jiang, Nitish Jindal, Zihan Li, Aaron Lin, Kyle Schmaus, Jay Tayade, Weicheng Zhao, Anupam Datta, Nathan Wiegand, Dimitris Tsirogiannis · 2025

Read Cortex AISQL: A Production SQL Engine for Unstructured Data Conference: ; ; by Paweł Liskowski, Benjamin Han, Paritosh Aggarwal, Bowei Chen, Boxin Jiang, Nitish Jindal, Zihan Li, Aaron Lin, Kyle Schmaus, Jay Tayade, Weicheng Zhao, Anupam Datta, Nathan Wiegand, Dimitris Tsirogiannis (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-07663.txt.

Description

Paweł Liskowski and colleagues at Snowflake describe Cortex AISQL, a database engine that answers questions about messy data with plain SQL. Users can mix normal relational queries with semantic operations that read text and other unstructured content. Production scale makes that blend hard, since semantic steps cost more and behave unlike joins or filters.

Costs and selectivity stay unknown while a query is compiled, so classic optimizers stumble. The engine redesigns planning and execution around those unknowns. Companies sit on mountains of documents beside their tables; a system that queries both together, declaratively, removes piles of glue code and custom pipelines.

Catalog record

CREATORPaweł Liskowski, Benjamin Han, Paritosh Aggarwal, Bowei Chen, Boxin Jiang, Nitish Jindal, Zihan Li, Aaron Lin, Kyle Schmaus, Jay Tayade, Weicheng Zhao, Anupam Datta, Nathan Wiegand, Dimitris Tsirogiannis
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-07663

Browse the full library · About the Tower of Babel