PRQL: A Modern Language for Data Transformation

This talk will introduce you to PRQL, a powerful data transformation language designed specifically for data analysts and engineers. PRQL combines the elegance of relational algebra with the usability of popular libraries like Pandas, Polars, and dplyr, offering a functional, pipelined query paradigm with modern ergonomics and expressivity.
Beyond seamlessly transpiling into SQL, PRQL also supports multiple SQL dialects. This ensures compatibility with any relational database and data warehouse and allows you to leverage your existing data infrastructure.

The talk will highlight PRQL’s expressivity with complex analytical queries, demonstrated through interactive examples in Python.

Vorkenntnisse

  • Knowledge of SQL is a prerequisite in order to appreciate how PRQL improves upon it.
  • Familiarity with data pipelines in Pandas, Polar, dplyr or dbt would be an advantage but is not required.

Lernziele

  • 1. SQL: a shark or dinosaur?

  • - What is great about SQL?

  • - Where does SQL fall short?

  • 2. Introducing PRQL: the Pipelined Relational Query Language

  • - Pipelined

  • - Relational


- Functional

- Orthogonal

- Ergonomic

3. Integrations: Use PRQL today!

- Online: PRQL Playground

- Databases: DuckDB & ClickHouse

- Python: prql-python and pyprql

- R: prqlr

- VSCode

- CLI: prqlc and pq

- and more ...

Speaker

 

Tobias Brandt
Tobias Brandt is a PRQL Core Contributor and an experienced professional in data science and data engineering with 20 years of experience in quantitative finance. He currently leads teams responsible for portfolio analytics, systems, and data at Argon Asset Management in Cape Town. Tobias has previously presented at industry conferences such as Subsurface Live and PyconZA.

data2day-Newsletter

Ihr möchtet über die data2day
auf dem Laufenden gehalten werden?

 

Anmelden