Grindr·Data Scientist·Technical Phone Screen
Jul 2026
Interviewed for a Data Scientist role at Grindr. Two meaty case questions, both heavily focused on experimentation and product analytics for their subscription and advertiser businesses. No fluff, just deep technical work from the start.
- A new advertiser optimization product is being evaluated but the pool of eligible advertisers is small. How would you design a credible causal study, including your estimand, hypothesis, randomization unit, experimental vs. quasi-experimental options, metrics, and power analysis?
- For the same advertiser experiment, what surprising or counterintuitive findings would you watch for, and how would you distinguish a real effect from noise?
- The CEO wants to lower the free-to-paid paywall from 100 profile views to 80. Walk through how you'd decide whether to ship it, including your objective, metrics, and experiment design.
- In the paywall threshold experiment, how do you handle the fact that only users who would have browsed past 80 profiles are actually exposed to the treatment? Why is a naive triggered-only analysis problematic?
- If the paywall change shows higher paid conversion but declining retention or engagement, what would you recommend and why?
“This one took a while to even orient to.”