Explain confounding with an Uber example

You are interviewing for a Senior Data Scientist role at Uber. 1) Define **confounding** in the context of estimating causal effects from observational data. 2) Give a **concrete Uber-related example** (not using age/sex/demographics) where a confounder would bias a naïve estimate. 3) Explain at least **two** practical ways you would detect/mitigate the confounding (e.g., in analysis design or modeling), and what assumptions each method requires.