Earlier in the week I had someone reach out to me to help them study for interviews to become a data engineer. She had a sample question related to sports and I asked to see her work. She showed me the question and an ERD she had created and I offered some help but I figured it may be useful to write about it.
Over the last few years I’ve done quite a few data engineering and data related interviews and the problem falls into a specific type of interview which I’d call the product sense interview. In my experience this is usually past the initial screening round and typically 45–60 minutes of SQL data modeling questions in an open ended format.
The purpose is to ask you a question like “You need to design the data model for Uber Eats (or Doordash). How would you do this?” then they ask you some metric related questions around it e.g. “Suppose you wanted to calculate the revenue per day by product category, write a query to do that with the tables you created.”. I think there’s usually supposed to be 3–5 SQL questions and they steadily get more difficult (basic select → aggregations → ranking & window functions).
In some cases it’s a take home and based on the company’s business model but I think this is less common. Anyway, I’ll explain a similar question and how to answer it.