Data types for data science.

Haskell is a strongly-typed, purely functional language, which in short means that many errors are caught by the compiler, resulting in fewer surprises when you run your code.
Your program is a data transformation pipeline, based on functions and expressions that can be composed unambiguously.
Also, Haskell is fast!

Anyone can participate in DataHaskell, regardless of data science or programming skill levels; we're always eager for new viewpoints and ideas!

DataHaskell is an open-source organization.

We are passionate about sharing ideas, code and projects. We focus on being friendly, completely open source, purely functional and keen on building and maintainting documentation. We intend to be integrators and ensure our library environment plays well with the rest of the Haskell ecosystem.


We are building a knowledge base of Haskell libraries and resources for data science, numerical computation and machine learning. Check it out here

GitHub Organization

The hub for all the code-related projects, from forks of other libraries to our own original tools. If you are planning on helping, or just want to take a look, this is your place


Our roadmap, with mid and long-term goals. You can check the status of what we are doing here.
See it here


A tech team for business and social innovation. Contributing with work to boost Haskell as a tool for the future of functional programming, data science and machine learning.

Ready to revolutionize data science?

It does not matter your skill level or work field, as long as you are interested in Haskell and/or data science you will find your place in dataHaskell.