dats'n'stats
Interview-based podcast on all things data science, entrepreneurship, statistics, machine learning, open source, and Python. Formerly PyData Deep Dive.
dats'n'stats
Kevin Systrom - After Instagram: Bayesian modeling of COVID-19 with PyMC3
•
Thomas Wiecki
•
Season 1
•
Episode 5
Kevin Systrom is the co-founder and former CEO of Instagram. In this episode we talk about his time after leaving the company, whether becoming a billionaire made him happier, how to decide what to work on, the Japanese concept of Ikigai, or a reason for being, and meditation.
We then dive into his current project -- rt.live -- a website which tracks the spread characteristics of COVID-19. The statistical model underlying the website is Bayesian and built using PyMC3. Kevin shares why he chose that approach and how he got started on PyMC3.
Show notes
- Kevin Systrom on Twitter and his blog
- Ikigai: The Japanese Secret to a Long and Happy Life
- The Mind Illuminated
- COVID-19 tracking: https://rt.live
- An Intuitive Guide to Bayesian Statistics
- PyMC3
- Statistical Rethinking by McElreath
- Bayesian Methods for Hackers by Cam Davidson-Pilon
Special thanks to all my patrons, specifically Jonathan Ng, Richard Craib, Daniel Gerlanc, and Ayan Doss.