Data Science Resources: Podcasts
As anyone who has achieved competence in their field knows, we didn’t wake up one morning knowing everything about our respective fields. We’ve read, studied, and put in many hours of work; and in the field of data science, this learning process is continuous to keep up with advancing technology. In addition to technical topics, we need to understand how business knowledge should inform our work, and how we can shape our work to inform business decisions. The relationship between data science and business should be an endless cycle of shared value.
One simple way to broaden our understanding of data science and its impacts on business and industry is by reading and listening to experts in the field discuss their work, how they approach it, and the tools they use. There are numerous such resources available. So many, in fact, that it can sometimes be overwhelming to decide which educational avenue to pursue first. I hope to review some books and websites in the future, but the focus of this blogpost is podcasts for continued learning in data science.
Below are some of my favorite (and not so favorite) podcasts focusing on data science and analytics.
Super Data Science
Hosted by Kirill Eremenko, Data Science coach and self-proclaimed “lifestyle entrepreneur”
As host, Eremenko interviews individuals in the field about their pathway to data science, their experiences, and their work. Eremenko’s intense enthusiasm is something people will either love or hate, but his passion for the field is undeniable. The episodes are somewhat hit or miss, but on the whole it’s interesting content with interesting guests. Topics vary widely, including episodes with information on industry and the economy, data science at Kaplan, insights from KDnuggets, concepts in robotics, importance of domain knowledge, and more. 240 episodes published and counting.
Hosted by Kyle Polich, MS in Artificial Intelligence, previously employed in teaching and analytic capacities at Thinkful, uStamp, and Dex One
The electronic intro music to this podcast makes you feel like you’re stepping straight into a video game, so I was hooked from the start. This podcast series is also built around interviews, but has a focus on outlining interesting problems the guests have faced and how he or she went about solving them. This problem-solution approach enhances understanding of applied data science. Another cool thing about Data Skeptic is that Polich includes frequent mini episodes which focus on explaining technical concepts that might be glossed over (with understanding assumed) in regular full-length episodes. Some memorable mini-episode topics were deep dive into p-values, descriptions of Bayesian statistics, and experimental design.
Hosted by Chris Albon, Data Scientist and Machine Learning Engineer; Vidya Spandana, Innovation Consultant, Entrepreneur, Technologist, and Presidential Innovation Fellow for Obama; and Jonathon Morgan, Software Architect, Data Scientist, and White House Data Science Advisor
Each episode begins with a casual introduction of the hosts describing the beverage they are drinking, usually of the alcoholic persuasion. These episodes have parental advisory flags, so probably best not to listen to with your aspiring young data scientist. What other podcasts sometimes deal with is bland personalities of interviewees. Partially Derivative (or Partially D, as they affectionately refer to themselves) avoids this problem by rarely having guests. The hosts are lively, entertaining, and well-informed, making this podcast a pleasure to listen to. They have an impressive ability to generalize detailed topics for wider application and foster a broader understanding of the topics addressed.
Data Crunch is a story-based podcast. It sweeps you along with the story, keeping you wondering what will happen next. Stories span the experiences of real-life people and focus on how data analysis provided the solution to problems, even if indirectly. This podcast is less technical than others, but enormously entertaining, providing example after example of how all of life revolves around data and patterns.
The first four episodes are a bit unusual. They focus on discussing the hosts’ fears of AI and what people can do to speak up and help frame the future. Recognizing these first four episodes may have missed the mark, the hosts labeled them -4 to -1 and resumed the series at episode 0 with a different tone. I get the impression that their knowledge and experience in the field is limited. Their discussions remain largely speculative, frequently taking dives into the philosophical and sociological aspects of tech, and the listener is left with very little confidence in the validity of the content. But if you’re interested in train wrecks or somewhat uninformed philosophical and sociological explorations, this podcast is for you. 70 + 5 episodes released.
Hosted by Katherine Gorman, former Public Radio Producer; and Neil Lawrence, Lead at Amazon Research Cambridge and Computational Biologist; formerly hosted by Ryan Adams, Professor of Computer Science at Princeton University
The material presented in this podcast is perhaps more academic than some, so listen with your glasses on! In early episodes, Talking Machines is hosted by Katherine Gorman, a novice in the field with deep analytical intuition and expert Ryan Adams; in later episodes computational biologist Neil Lawrence assumes Ryan’s place. The interaction between the expert knowledge and the enthusiastic curiosity and unique perspective of the novice give an energy to the show that is refreshing. Topics covered include the social scene of data and analytics research, descriptions and recommendations of books and conferences in the field, trends in industry, machine learning across disciplines, and more technical topics such as the use of Jupyter Notebooks, Probabilistic Programming, and more.