March Interlude – Data Science Podcasts I Follow

I commute (via car) from NY to South Carolina on a regular basis. That 13 hour drive lends itself to great opportunity to listen to new music, get lost in an audiobook or be entertained & informed by podcasts.

imageThe only app I’ve ever used for podcast management is PodCast Addict. I enter in the URLs of the podcasts I want to follow and it notifies me when new episodes are available. Best of all, I can download those podcasts to my phone over a wifi connection and listen to them offline while I travel.

Below are the podcasts I follow religiously, having not missed an episode for months. I list these in no particular order. Click on the podcast logo to be taken to their site.


imageBack in July I wrote about an incredible book called The Data Science Handbook, a captivating collection of interviews with 25 data scientists from a variety of backgrounds who are out there, every day, doing it!

Renee Teate has implemented a podcast version of this book (although she has no association with it to my knowledge). To date, she’s conducted 7 interviews with a diverse group of data scientists and she’s just recently “complained” that’s she’s nowhere near done with the original list of individuals she brainstormed when she had first developed this idea. One of the most captivating interviews was with Safia Abdalla (@captainsafia), a student at Northwestern University who’s already leading Python Data Science groups and speaking (inter)nationally at conferences!

Podcast frequency: Every other week

Follow Rene on Twitter (@BecomingDataSci).


image

With Data Skeptic, you really get two podcasts in one. On one hand, you get mini-episodes – brief, non-technical discussions between Kyle (the “expert”) and Linh Da (the “non-expert”) on pure Data Science topics like R-squared, multiple regression, the Bonferroni Correction, etc. And I truly mean they are non-technical. Kyle explains these concepts with metaphors, analogies, similes and perhaps one day, in song. Sometimes it’s refreshing to pull your head out of a book of math equations and hear something explained like you’re a 5-year-old (or at least like you’re an undergraduate student).

The other episodes are longer in nature and feature Kyle interviewing various experts on Data-Science-in-the-world topics. Recent topics include: “Models of Mental Simulation”, “Auditing Algorithms” and one of my favorite episodes: “Detecting Pseudo-profound BS”.

Podcast frequency:  Weekly, alternating between mini-episodes and longer length episodes.


image

Partially Derivative is the frat party of Data Scientist podcasts. With 25% of each episode devoted to discussing specialty beers and another 25% devoted to begging listeners for beer donations, the remaining 50% involved high-spirited discussions about Data Science in the news ranging from controversial journal publications to politics.

The three hosts (Jonathan, Vidya and Chris) are founders of the online analytics engine, Popily, so they’ve been out there risking their necks to try to make a living off of Data Science. The least you can do is drop in and listen to their data science bar conversations.

Podcast frequency: Whenever they feel like it: sometime weekly, sometimes biweekly, sometimes they miss February.


image

On the more serious side, Katie and Ben cover data science topics (usually requested by listeners) in more depth.  Each episode focuses on a single topic such as Neural Nets, p-Values and natural language translation (Yiddish, anyone?).

Katie is the anchor to this podcast and with a background in experimental physics (having worked at the Large Hadron Collider at CERN) I find her expansion into data science to be especially interesting. Recently, Katie announced that she will be spending a significant time volunteering work on the White House’s recent National Cancer Initiative. I applaud her efforts and accomplishments.

Ben seems like a nice guy too.

Podcast frequency: Weekly (sometimes more than 1 episode in a given week)


image

Not So Standard Deviations is a podcast by Roger Peng (of Coursera’s JHU Data Science Specialization fame) and Hilary Parker (Senior Data Analyst at Etsy) where they discuss a wide variety of miscellaneous, unrelated topics.

It’s a light-hearted conversation between the two, where they dive into “controversial” topics like the deficiencies of MS Excel as a data science tool, why Roger doesn’t use Netflix and the R package cat, a God-as-my-witness, working R package  devoted to furry & purry cat functions.

Podcast Frequency: Every other week


image

The O’Reilly Data Show Podcast dives more into the Big Data side of things, covering the various architectures and movers & players in the terabyte world.

Each episode features a lengthy interview with a key player in the big data world. Recent episodes spotlighted M.C. Srivas, co-founder of MapR (which holds various speed records for data querying), Fang Yu, co-founder and CTO of DataVisor about using Apache Spark to predict security attack vectors in real-time and Eric Colson, former VP of data science and engineering at Netflix, about the need for human input in the data-science world.

Podcast Frequency: Every other week


image

Signal is NOT a Data Science podcast and I do not recall how I ended up subscribing to it. But it is among my favorites.

It focus exclusively on the drug industry: Drug trial, testing procedures, pricing, availability, competition, business acquisitions, investments, politics, etc.

This is the most professionally produced podcast on this list and is an absolutely riveting look at what goes on in the world of prescription drugs. The most recent episode, “How much are we willing to pay for cures” takes you through the history of treatments for hepatitis C, which is now curable with a high success rate, but at a price.

Frequency: Every other week.


Do you have any podcasts among your favorites that you’d recommend? I’m eager to hear your suggestions below in the comments.