A programming language for scientific machine learning and differentiable programming

The Data Exchange Podcast: Viral Shah on the evolution and maturation of Julia.


SubscribeiTunesAndroidSpotifyStitcherGoogle, and RSS.

In this episode of the Data Exchange I speak with Viral Shah, co-founder and CEO, Julia Computing. Along with his Julia language co-creators, Viral was awarded the 2019 Wilkinson prize, for outstanding contributions in the field of numerical software. I first tweeted about Julia at the beginning of March 2012  after seeing Jeff Bezanson give a talk in Stanford. I’ve dabbled with it here and there, but have never used it for a major project. Over the past few years, Julia continued to add packages at a steady pace and the package manager is really quite impressive and solid.  We spent much of the podcast discussing the state of Julia, Julia 1.5, the Julia ecosystem and community.

Download the 2020 NLP Survey Report and learn how companies are using and implementing natural language technologies.

If your application requires complex computations and you have strict performance requirements, Julia is definitely a solid candidate. With that said,  the rise of deep learning has made Python the dominant language for ML as many of the popular libraries targeted Python from the get-go.

We also discussed Viral’s 2016 book –  Rebooting India: Realizing a Billion Aspirations – which he co-wrote with Nandan Nilekani, co-founder of Infosys Technologies.

Subscribe to our Newsletter:
We also publish a popular newsletter where we share highlights from recent episodes, trends in AI / machine learning / data, and a collection of recommendations.

Related content:


[Image: Mandel zoom 12 satellite spirally wheel with julia islands from Wikimedia]