Explain ETL schema changes and ensure integrity
Company: Rbcroyalbank
Role: Data Engineer
Category: Behavioral & Leadership
Difficulty: easy
Interview Round: Technical Screen
## Behavioral / Data Engineering Discussion Prompts
You’re interviewing for a **Data Engineer** role. Answer the following prompts with concrete examples from your past projects (use the STAR framework where helpful):
1. **New skills from your Master’s program**
- What specific skills did you gain that made you better at building/operating data pipelines?
2. **Handling schema differences in ETL**
- In an ETL/ELT pipeline, how do you handle **schema differences** between source and target (e.g., missing/extra columns, type changes, renamed fields, nested JSON evolving over time)?
- When do you choose strict schema enforcement vs flexible evolution?
3. **Ensuring data integrity**
- How do you ensure **data integrity** end-to-end (accuracy, completeness, consistency, and timeliness) across ingestion → transformations → warehouse tables?
- What checks do you implement (and where), and how do you respond to failures?
4. **Pipeline difficulties**
- Describe one difficult pipeline problem you encountered (e.g., late-arriving data, duplicates, backfills, scaling, flaky upstreams, partitioning, idempotency, cost blowups, SLA misses).
- What was the root cause, what did you change, and what did you learn?
Overview: This question evaluates proficiency in handling ETL schema evolution, maintaining end-to-end data integrity (accuracy, completeness, consistency, timeliness), pipeline reliability, and leadership in incident response, and is commonly asked to gauge experience managing schema drift, enforcing data quality, and owning production pipelines.