Tag
You don't need to be an ML engineer to make good architecture decisions with AI. The minimum concepts every architect should grasp before putting a model into production.
CQRS isn't just "separating reads from writes" — it's accepting that, for an instant, different parts of your system will disagree about the truth. When that trade-off is worth it, and when it's over-engineering.
Code tells you what the system does. It almost never tells you why it was decided to be built that way. Architecture Decision Records are the simplest way I know to not lose that second part.
We explore the most effective architectural patterns when designing scalable solutions on AWS.
Redundancy is not the same as resilience. Notes on designing architectures that absorb failure instead of avoiding it, using ISO/IEC 25010 as a design language.
