Great Data Science and Machine Learning Podcasts


This post was originally published by Sadrach Pierre, Ph.D. at Towards Data Science

The U.S. Bureau of Labor statistics estimates that 11.5 million new data science jobs will be added by 2026. This projection suggests that the field of data science will continue to rapidly grow with the high demand for these positions in industry. For those just starting out in the space or considering a transition from another industry, podcasts are useful for developing an understanding of data science and machine learning in industry and research. In this post, I will discuss three of my favorite podcasts that discuss the data science field in research and industry. These podcasts also suggest learning resources for those how are just starting out.

Let’s get started!


The SuperDataScience podcast is hosted by Kirill Eremenko, who is a data science coach and entrepreneur. SuperDataScience features many leading data scientists and data analysts who provide insights into how to establish a successful career in data science. This podcast is a must for anyone who wants to better understand the industry and continue to educate themselves in the field. One of my favorites is SDS 391: Data Science Campfire Tales with John Elder. In this episode, Kirill and John Elder discuss mathematical concepts such as calculus, statistics and resampling. They also discuss the importance of domain knowledge, thoughts on neural networks, thoughts on the future of data science and much more. I also enjoyed SDS 373: TensorFlow and AI Learning for Developers. Here, Kirill sits down with Laurence Moroney to discuss TensorFlow and how developers’ can use it to progress their careers in data science. I also recommend SDS 379: Maelstrom, Chaos, and Mayhem: Guiding Your Data Science Career Path. Here, Kirill sits down with Christopher Bishop who has outlined a framework to help those starting out in data science identify their passion in the field.

Lex Fridman Podcast

In this podcast, MIT AI researcher Lex Fridman has conversations about AI, science, technology and more. Specifically, much of the content revolves around deep learning, AI robotics, computer vision, artificial general intelligence, computer science, and neuroscience. In contrast to the SuperDataScience podcast, this podcast focuses much more heavily on broad concepts such as intelligence and consciousness. This is a great podcast for those interested in the broad applications of machine learning methods as they are employed in artificial intelligence systems as well as those interested in how the brain works. I highly recommend #106 — Matt Botvinick: Neuroscience, and AI at DeepMind, where Fridman sits down with Matt Botvinick to discuss neuronal mechanisms of the brain as it relates to learning. I also recommend #101 — Joscha Bach: Artificial Consciousness and the Nature of Reality, where Fridman and Bach discuss the workings of the human brain, autonomous robots, discontinuity of existence, and many more interesting philosophical musings. Finally, I recommend #81 — Anca Dragan: Human-Robot Interaction and Reward engineering. On this podcast, Fridman and Anca discuss how robots as agents in the world perform tasks while navigating objects and people.

Data Skeptic

The Data Skeptic podcast is hosted by Kyle Polich and it features interviews and discussions around data science, statistics, machine learning, and artificial intelligence. This podcast is much more research focused where many episodes involve discussions of recent papers in the space of machine learning. I recommend Interpretable AI in Healthcare, where Polich sits down with Jayaraman Thiagarajan to discuss his paper Calibrating Healthcare AI: Toward Reliable and Interpretable Deep Predictive models. In this podcast, Polich and Thiagarajan discuss formalizing the process of explaining model predictions in healthcare to doctors. I also enjoyed Self-Driving Cars and Pedestrians, where Polich sits down with Arash Kalatian to discuss how to decode pedestrian and automated vehicle interactions using virtual reality and interpretable deep learning. Finally, I recommend Adversarial Explanations, where Polich sits down with Walt Woods to discuss his paper Adversarial Explanations for Understanding Image Classification Decisions and Improved Neural Network Robustness. Here, Polich and Woods talk about the concept of adversarial explanations which involve heat mapping at the intermediate layers of neural networks.


To summarize, in this post we discussed three great podcasts that discuss machine learning and data science. First we discussed SuperDataScience, which focuses on career advice for budding data scientists. Next we talked about the Lex Fridman Podcast, which tackles many interesting questions around intelligence and consciousness. Finally, we discussed Data Skeptic, which features many discussion around frontline research in the space of machine learning. I hope you found this post interesting/useful. Thank you for reading!

Spread the word

This post was originally published by Sadrach Pierre, Ph.D. at Towards Data Science

Related posts