OLTP (Online Transaction Processing): systems that run the business day-to-day. Lots of small reads/writes, strict consistency, normalized schemas. Examples: order DB, payments DB.
OLAP (Online Analytical Processing): systems for analysis. Large scans, aggregations, historical trends. Columnar warehouses and lakehouses. Examples: Snowflake, BigQuery, Redshift, Databricks SQL.
OLTP: UPDATE orders SET status='paid' WHERE id=42 (point lookup) OLAP: SELECT day, SUM(amount) FROM fct_orders GROUP BY day (scan + agg)
How they connect
CDC/ETL/ELT moves OLTP changes into OLAP models so analysts do not hammer production DBs.
Interview tip: "OLTP = run the app; OLAP = analyze the business." Mention different schemas (3NF vs star) as a follow-up.