How an MIT Research Project Became the Julia Programming Language

In 2009, a group of researchers frustrated with the programming languages available for complex mathematical and statistical work started a research project at MIT to build a new one. The languages available to scientists at the time were rigid and slow: if a researcher built something that worked, they often had to rewrite the entire program in another language just to make it run faster.

That project became Julia, a free, open-source programming language designed for scientific research, data analysis and modeling complex systems such as jet engines, drugs, financial markets and robots. It grew from a research project into a lab at MIT and then into a company, JuliaHub, co-founded by Julia's creators: Viral Shah, MIT mathematics professor Alan Edelman, Jeff Bezanson and Stefan Karpinski.

Julia is now used by more than 1 million people across thousands of companies and universities, for tasks ranging from modeling atoms and semiconductors to neural networks, race cars and airplanes, and by astronomers imaging black holes. Its speed comes from "just-in-time compilation," which compiles code differently depending on the type of data being used, making it faster and more flexible than other numerical programming languages.

The MIT researchers announced Julia in a blog post in 2012, three years after starting the project in what became MIT's Computer Science and Artificial Intelligence Laboratory. In April, JuliaHub launched Dyad 3.0, the latest version of a platform that lets engineering teams direct AI agents through physics simulations, safety analyses and quality controls while designing complex physical systems such as rockets, heat pumps and satellites.