VARIANCE

VARIANCE
Podcast Description
Variance explores the journeys of Data scientists, Product owners, Econometricians, Consultants specializing in the field of Experimentation and Observational Causal inference. Hosted by Pritul, a Data scientist who has spent 11 years scaling experimentation methods, infrastructure, and programs at scale at companies like Apple (Apple Podcasts), Peacock TV (streaming), ebay (ecommerce), and Yahoo (media), solving unique challenges of statistics and scalability.
Each episode is with a new guest where topics like debates in statistical methods, history, unique perspectives.
Podcast Insights
Content Themes
The podcast delves into topics related to experimentation and observational causal inference, discussing statistical debates and personal journeys. Episodes such as Statistical Pitfalls in Tech Experimentation focus on challenges faced in experimentation methods at major tech firms, while the inaugural episode features Ishan Goel, exploring the transition from computer science to data science.

Variance explores the journeys of Data scientists, Product owners, Econometricians, Consultants specializing in the field of Experimentation and Observational Causal inference. Hosted by Pritul, a Data scientist who has spent 11 years scaling experimentation methods, infrastructure, and programs at scale at companies like Apple (Apple Podcasts), Peacock TV (streaming), ebay (ecommerce), and Yahoo (media), solving unique challenges of statistics and scalability.
Each episode is with a new guest where topics like debates in statistical methods, history, unique perspectives.
In this special second episode, the roles reverse as I share some stories from my extensive experimentation background across tech giants like eBay, Apple, and Yahoo. This is a cut from my first episode with Ishan where Ishan asked me about my experience.
Discover the statistical challenges that plagued major experimentation platforms – from misaligned alpha thresholds and rampant false positives to the complexities of handling extremely skewed revenue metrics. Learn how I rebuilt trust in A/A testing through blind studies, applied multiple comparison corrections, and navigated the challenges of KPI engineering for metrics with distributions spanning from $1 to $8,000. This candid behind-the-scenes look into real-world experimentation reveals why understanding statistical concepts is crucial for accurate decision-making in tech. Perfect for data scientists, analysts, and product managers working with experiments and causal inference at scale.

Disclaimer
This podcast’s information is provided for general reference and was obtained from publicly accessible sources. The Podcast Collaborative neither produces nor verifies the content, accuracy, or suitability of this podcast. Views and opinions belong solely to the podcast creators and guests.
For a complete disclaimer, please see our Full Disclaimer on the archive page. The Podcast Collaborative bears no responsibility for the podcast’s themes, language, or overall content. Listener discretion is advised. Read our Terms of Use and Privacy Policy for more details.