How should secure error handling be designed for production?

For production, map expected failures to stable error contracts, use generic messages for unexpected faults, attach correlation IDs, log details securely, avoid sensitive data in logs, and make authorization failures indistinguishable where resource existence is sensitive. Add automated tests and observability around the critical behavior, document ownership and failure handling, and review the design when traffic, dependencies, or security requirements change.