In this episode of De Databar (Dutch), I talk about the background of Polars, what makes it fast, and what went into writing the O’Reilly book with Jeroen Janssens.
Topics covered:
- History of Polars: How Polars started at Xomnia and grew into one of the fastest DataFrame libraries available.
- Performance advantages: Why Polars outperforms Pandas on most workloads and how the Lazy API contributes to that.
- Competing frameworks: How Polars compares to Spark, DuckDB, and other modern data tools.
- Polars Cloud: What distributed Polars execution looks like and where the project is headed.
- LLMs and Polars: How large language models handle Polars code today and what that means for developers.
- The book: What drove us to write a comprehensive guide and what readers can expect.