Course Test

Question 1

What is the gap between an infrastructure 'intent' and a 'topology'?

Select one answer

The intent is written in YAML; the topology is written in natural language

The intent states what is needed; the topology names every component, service, data dependency, and networking rule required to deliver it

The intent is created by the LLM; the topology is created by the engineer

There is no gap - a well-written intent is equivalent to a topology

Question 2

Which Kubernetes primitives should you map out before handing a brief to an LLM, to use as a completeness checklist for the output?

Select all that apply

Deployment or StatefulSet per workload

Service per workload

Helm chart version per dependency

NetworkPolicy per namespace or isolation boundary

PersistentVolumeClaim per stateful workload

Question 3

In the co-design loop, what should you do when you find a problem in the Kanvas canvas after an LLM proposal?

Select one answer

Fix the problem directly in the Kanvas canvas and deploy

Discard the design and start over without LLM assistance

Compile a correction list and return it to the LLM as a correction brief, then review the revised proposal

Skip the problem if it appears minor and proceed to deployment

Question 4

What does an 'Exclusions' section in an infrastructure brief accomplish?

Select one answer

Lists the components the LLM must include in the design

Prevents the LLM from adding unauthorized components by explicitly stating what is out of scope

Documents components that were considered but are deferred to a future iteration

Reduces the context window size by removing irrelevant information

Question 5

Which of the following are resource hygiene checks you should perform when reviewing an AI-proposed design?

Select all that apply

Every container has resources.requests.cpu and resources.requests.memory set

Every container has resources.limits.memory set

Every Service uses the LoadBalancer type for external access

PodDisruptionBudgets are present for availability-sensitive workloads

HPA min and max replica counts are consistent with brief availability requirements

Question 6

When during a design review should you update the infrastructure brief rather than sending a correction list to the LLM?

Select one answer

When the design has more than three corrections needed

When you discover during review that you are inventing requirements that were never in the brief

When the LLM proposal is more than 100 lines of YAML

When the design passes the completeness check but fails the security check