An example from practice
- Context
- A software vendor had its multi-tenant microservice platform on .NET, EF Core and PostgreSQL reviewed for code quality, performance under load and security-relevant details.
- Finding
- Blocking .Result calls inside async database code, a random generator that wasn't cryptographically secure, and avoidable memory allocations in hot paths such as string building and logging.
- Outcome
- Every recommendation came with a benchmark and a code sample. Measured: up to 85% less memory per call and hot paths running up to almost five times faster.

