These come up again and again in interviews for analyst and analytics engineer roles that use BigQuery: a few to write, a few to explain. For the SQL ones, interviewers look for BigQuery's own tools: QUALIFY, UNNEST, SAFE_DIVIDE, wildcard tables. For the others, the short answer and the reason behind it.
Read every explanation, even when you're right: it's often the follow-up question.
State assumptions before choosing the query. Ask whether a ranking should keep ties, whether missing dates should appear and whether an aggregate measures events or people. Several SQL formulations can produce the same correct result; use the one you can explain. Each runnable answer is checked on changed data. The operational questions test a stated constraint, rather than a memorized vendor slogan.