A/B Testing Interview with a Google Data Scientist

I'm back with Sandeep, an ex-Google data scientist and Ph.D. student at the University of Minnesota, who answers an A/B testing mock interview question asked by several companies like Airbnb, Asana, and Square.
Here's the interview question we're tackling: www.interviewq...
Check out Sandeep's website: sandeepgangara...
Want to be featured in the next mock interview video? Apply here: airtable.com/s...
👉 Subscribe to my data science channel: bit.ly/2xYkyUM
🔥 Get 10% off your next data science interview prep: www.interviewq...
❓ Check out our data science interview course: www.interviewq...
🔑 Get professional coaching here: www.interviewq...
🐦 Follow us on Twitter: / interview_query
Looking to hire top-tier data scientists like Sandeep? Outsearch.ai leverages AI to streamline your hiring process, matching you with experienced professionals skilled in A/B testing and other crucial data science techniques: www.outsearch....
Relevant Links:
A/B Testing and Statistics Interview Questions: www.interviewq...
More from Jay:
Read my personal blog: datastream.sub...
Follow me on Linkedin: / jay-feng-ab66b049
Find me on Twitter: / datasciencejay

[@iqjayfeng]

Check out our new statistics and a/b testing course here: www.interviewquery.com/courses/data-science-course/lessons/statistics-and-ab-testing

[@prithviprakash1110]

Thanks for the great video Jay. I just wish there was a portion in the end of the video dedicated towards Sandeep receiving some feedback for us viewers to understand what he got right and wrong.

[@candidlyvivian]

Thanks for sharing Jay! As a fellow data scientist (and fellow STEM youtuber) in Australia, can say that I've actually been asked very similar design thinking type questions in my interviews!

[@aravindkramesh]

How do I become a product analyst at google?

[@faridsaud6567]

@@aravindkramesh bro has no chill.

[@jeremynx]

Thank you very much, Sandeep, for your detailed and clear answer. A big thanks to Jay as well for the great questions.

[@iqjayfeng]

You're always welcome!

[@MissAngelFaace]

This is great. Thanks so much for posting, Jay and Sandeep!

[@chiranjeeveemohapatra]

This post is little old but I would still add points here: We could do something like treatment I and II groups testing $5 and $15 that will capture both less than and greater than 10. Also the last question was about novelty effect. So mentioning that should help.

[@pavan5208]

Hi found this really insightful. Thank you. anyone please help me to answer this question which I got recently: "what is the right size for an experiment? /How do you decide sample size for an experiment?/ What is the right size for an experiment?" I understood we need to decide based on practical significance,, but can you please help me to understand with an example? TIA!

[@dolusolana2127]

Keep them coming, Jay.

[@anibalgonzalez7990]

Good stuff!

[@aravindkramesh]

*I wish I could get a Product Analyst Internship at Google.*

[@fmxy]

I first thought of people not wanting to provide a payment method or bank account and being put off because of that

[@pooh1989kolkata]

They can give an amazon voucher or something similar without the need of a user's account details. Also, this could be covered in the entire experience part that he is talking about for the treatment group but good point! :)