Databricks Interview Question

Can you explain how CI/CD works in a data engineering context, and how you'd implement it for your Databricks notebooks or pipelines?