Course Test

Please answer the questions below to complete the test.

Question 1

Which of the following best describes the 'lost in the middle' problem with large context windows?

Select one answer

Models refuse to process context windows larger than a fixed token limit

Models attend poorly to content buried deep in the middle of a long context

Context windows lose tokens from the middle due to compression

Retrieval systems fail to return the middle chunks from a document

Question 2

An agent needs to know the current number of ready replicas for the payment service in the production namespace. Which approach provides the most accurate answer?

Select one answer

Inject a pre-fetched cluster snapshot taken at session start

Rely on the LLM's training data for typical replica counts

Execute a runtime tool call to kubectl get deployment

Query the Meshery Catalog for the service definition

Question 3

When building a retrieval knowledge base from runbooks, which chunking approach is most appropriate?

Select all that apply

Split each runbook into one chunk per procedure, with a header naming the procedure and service

Attach metadata including source file, last-modified date, and relevant service or namespace

Store each entire runbook as a single chunk to preserve all context

Split on every paragraph boundary with no metadata

Question 4

What command verifies that Meshery is running correctly before an agent begins any retrieval or design import operations?

Select one answer

mesheryctl system start

kubectl get pods -n meshery

mesheryctl system check

mesheryctl registry list

Question 5

Which of the following are valid safety requirements before an agent executes a write operation on a production cluster?

Select all that apply

Require explicit human approval before any kubectl apply or design import

Run a dry-run and surface the output for review before executing

Evaluate the proposed change against OPA or Meshery policy constraints

Allow the agent to apply changes automatically if tool call accuracy exceeds 80%

Question 6

What is the primary purpose of a 'golden task set' in the context of evaluating an infrastructure agent?

Type your answer below