Algorithm · Boston Consulting Group · Hard
You are given two CSV files of user transactions and user profiles. Write idiomatic, production-ready pandas code that turns them into a per-user feature snapshot. Do not use groupby.apply or explicit Python loops over rows, and assume the data can be 100M+ rows. transactions.csv Columns: txn_id user_id ts_iso: ISO 8601 timestamp with a time zone amount: decimal USD; refunds are negative merchant_cat type: one of purchase, refund, chargeback updated_at: last write wins…
Checking your access…