advanced

Data engineering basics

Understand OLTP versus OLAP, ETL/ELT, batch and streaming pipelines, warehouses, data lakes, quality, lineage, and analytics boundaries.

Data engineering basics interviews separate OLTP serving from OLAP analytics and cover safe schema migration, backups, replication, CDC, ETL/ELT pipelines, and data quality or lineage. FullStack candidates should place these in architecture without over-building every app.

Subtopics: OLTP vs OLAP, migrations, backups, replication, CDC, ETL/ELT pipelines, data quality and lineage.

On interviews: sketch how transactional data reaches a warehouse, what breaks during schema change, and how you would prove a backup restore works.

Common pitfalls: analytics queries on production OLTP; migrations without expand-contract; pipelines without idempotency; no owner for data freshness.

The trade-off is balancing analytics insight against pipeline complexity, cost, and governance overhead.

Checklist:

Separate OLTP paths from OLAP workloads.
Version migrations with backward-compatible deploys.
Test restores, not just backup jobs.
Make pipelines idempotent with observable freshness.