How many tables were you responsible for, roughly, and how big was the biggest one?

I was responsible for a data warehouse that had roughly 120 tables spread across three schemas. The largest table was a fact table that stored click‑stream events and contained about 250 million rows, which equated to roughly 15 GB of compressed Parquet data in Snowflake. We used partitioning on the event date and clustering on user_id to keep query performance high. The schema design followed a star‑schema pattern, with dimension tables for users, products, and time. We regularly monitored table growth using Snowflake’s INFORMATION_SCHEMA tables and set up alerts when a table exceeded its expected size. This helped us plan for scaling and cost optimization, such as moving infrequently accessed data to a cheaper storage tier.