Simon Aubury

A quack of all trades: DuckDB, a versatile analytical database to keep in your toolkit

DuckDB is a fast in-process analytical database that enables you to load, clean, transform, summarise, and export large datasets with ease. It’s simple to use—just import DuckDB within your notebook or application code and you’re ready to go—no database server required! DuckDB is notable for its versatility across a wide range of analytical data use-cases. Originally designed with data scientists and data analysts in mind, enabling them to scale and supercharge their analytical workflows, DuckDB is also being increasingly adopted by data engineers and software engineers, using it as a lean building block for operational data infrastructure and interactive data products. In this talk, we will first demystify DuckDB by covering its core features and comparing and contrasting it with other databases and data processing tools. They’ll then take you on a whirlwind tour of a range of different use-cases that DuckDB is a great fit for, from different types of data analysis to data pipelines, data lakehouse patterns, and interactive data apps. You’ll learn why a data engineer and a data scientist finally agreed on a shared tool and why you might want to consider adding DuckDB to your toolkit.