Architecture growth demo

⚔ Creature Forge ⚔

Start simple. Add segments. Add requirements. Watch where the complexity lands.

1. Who should understand the backend?

Management version:

Explain it for:

Product growth simulator

2. Let the product grow

Each step represents a perfectly reasonable requirement. The backend REST services stay in place.

Stage1/ 6

Dependency picture

3. What the UI is coupled to

Complexity meter

What changes as we scale?

LOW

1backend endpoints UI knows

0routing rules in UI

2REST services reused

1consumer change surfaces

Live request

4. Run the same use case

Choose a segment. The more specialized the segment, the more backend pieces it needs as product requirements grow.

—UI business requestsRun an action.

—routing decisions in UIRun an action.

—REST calls underneathSame backend investment can be reused.

—consumer integration pointsPlaces a consumer must know can change/fail.

5. What the UI has to know

Run a request to see the difference.

6. Final object the UI receives

The screen can receive the same domain-shaped result either way.

No creature yet.

7. Follow the request

Run an action and the request path will appear here.

The point

8. This is not about saving one HTTP call

GraphQL does not eliminate complexity. It moves backend composition complexity away from every consumer and into a controlled API layer.

Direct REST

REST BFF

GraphQL

Warehouse analogy: Direct REST gives every customer a warehouse map. A REST BFF gives them a staffed pickup counter with a fixed menu. GraphQL gives them a service desk where they can ask for exactly the fields they need. All three can use the same warehouse underneath.