Polars

Polars began as an open-source project created by Ritchie Vink to provide a faster, more expressive approach to structured-data manipulation. Rather than adapting existing DataFrame architecture, the project was built from the ground up around performance, multithreading, vectorized and columnar execution, and fine-grained control over memory and execution. Its Rust core and Apache Arrow memory model helped establish Polars as one of the fastest DataFrame libraries for single-machine workloads.

As adoption grew among data engineers, analysts, data scientists, and trading firms, Polars Inc. was founded in 2023 to provide sustained engineering support and commercial products. Ritchie Vink subsequently invited Chiel Peters to join him as a co-founder. The company now develops Polars Cloud and distributed query capabilities that allow the same Polars code to run from a local machine to production clusters in cloud or on-premise environments. The open-source project remains MIT-licensed and community-driven, while the company supports it through enterprise software, distributed execution, and related services.