The Data Exchange Podcast: Dean Wampler on Ray, distributed systems, and Scala and Python.
In this episode of the Data Exchange I speak with Dean Wampler, Head of Developer Relations at Anyscale, the startup founded by the creators of Ray. Ray is a distributed execution framework that makes it easy to scale machine learning and Python applications. It has a very simple API and as someone who uses both Python and machine learning, Ray has been a wonderful addition to my toolbox. Dean has long been one of my favorite architects, speakers and teachers, and we have known each other since the early days of Apache Spark. He has authored numerous books and is known for his interest in Scala and programming languages, as well as in software architecture.
Our conversation spanned many topics, including:
- What is Ray and why should someone consider using it?
- The first Ray Summit (May 27-28 in San Francisco)
- Dean’s first impressions of Ray, and his journey from Scala to Python.
- An update on Ray’s core libraries, Ray on Windows, and distributed training with Ray.
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.
- Announcing Ray Summit, May 27-28, 2020
- Dean Wampler on “Ray for the curious”
- Dean Wampler on “Ray Tips and Tricks, Part I — ray.wait”
- Dean Wampler on “Distributed Python for Massive Scalability”
- Rajat Monga: The evolution of TensorFlow and of machine learning infrastructure
[Photo from Dean Wampler (Art on the Mart), used with permission.]