Airbnb DS Hiring Committee Member Interview
I’m in the last leg of an Airbnb Staff DS role (not analytics). I’m meeting with a DS lead that’s on the hiring committee for leveling. What should I expect here? (Can’t share my TC, this is specific enough and hiring for new roles is limited enough that it would be clear to anyone in the interview process that I’m the one asking this question) #datascience #datascientist
Good luck with your interview. I am also targeting the Sr. role in DS (not analytics), but lacking the skills. If you have a minute, could you please advise on what topics/aspects to prepare for the interview, resources and what kind (just type, not specifics) of questions to expect in interviews. Thank you
Are you doing the causal inference or ML path? I did CI: For prep, start with reading thoroughly the Wikipedia page for the following topics: Hypothesis Testing Statistical Power Multiple Hypothesis Correct OLS Regression Instrumental Variables Propensity Score Matching Matching Designs Difference in Difference Regression Discontinuity Synthetic Control Next read the following: *For foundational statistics: All of Statistics: A Concise Course in Statistical Inference (Springer Texts in Statistics) https://a.co/d/6zpUlgU. It is very technical but give you a solid foundation for stats *For experimentation: Trustworthy Online Controlled Experiments https://a.co/d/1Bnb2eg. For econometrics oriented causal inference: Mostly Harmless Econometrics: An Empiricist's Companion https://a.co/d/65FgQvl. Also read up on synthetic controls for full coverage. *For a graph/DAGs oriented approach to causal inference: Counterfactuals and Causal Inference: Methods and Principles for Social Research (Analytical Methods for Social Research) https://a.co/d/3QT3b0L If you’ve never been to exposed to causal inference, do Introductory Econometrics: A Modern Approach (MindTap Course List) https://a.co/d/aUK1Q3I (If you are short on time, focus on the ones that I stared, read those before the rest and you will have most of the knowledge you need to clear these interviews) Next practice implementing basic statistical test from scratch (i.e. only using the statistical distribution related functions in scipy and numpy). For example, compute a pvalue for a hypothesis test without using any of the python or R packages that do all the work for you. While I’ve never been asked to do this in an interview, I found that doing stats101 methods manually is the best way to solidify a deep understanding of the concepts. In fact, a deep understanding of the core methods puts you in a better place than memorizing the names and high level applications of a large number of methods. Practice using all the methods I mentioned in the first part. When you use the statistical packages, make sure you understand every aspect of the outputs and inputs. Again, you might not be asked this directly in an interview but it will give you the confidence and deep understanding of these methods that you need to answer interview questions quickly and thoroughly. I finished my statistics technical screens in half the time allotted b/c I understood the basics very well.
Thank you very much. That is very helpful. I am interested in doing the ML path (which I am in currently but with less skills). I remember a different post from you about the prep material and about casual inference resources. Do you mind sharing your YoE ? Thanks again for all the great details on prep.
🤦♂️ smart people. Wouldn’t your last statement in the post reveal who you are.
Yeap! I think risking letting them know I’m asking for tips on this interview is fine. What’s not fine is sharing my TC before salary negotiations even starts.