Claude Certified Architect – Foundations Practice Question 21: Orchestrator Workers Subtasks
Claude Certified Architect, Foundations · Test 3 · question 1 of 10
A team is building an assistant that implements feature requests across a large monorepo. Depending on the request, the change may touch two files or forty, and nobody can say in advance which files or how many edits are needed until the codebase has been inspected. The team wants the work split up and run by separate Claude calls, then merged into one coherent result. Which design fits best?
Show answer and explanation
Answer:
Orchestrator-workers is built for tasks where the subtasks cannot be predicted ahead of time, such as changes to an unknown number of files. The orchestrator breaks the task down dynamically, delegates to workers, and merges what they return.
- A is wrong: sectioning splits a task into independent subtasks that are known up front; the number of prompts is fixed in code, so it cannot adapt to two files versus forty. - B is wrong: prompt chaining is a fixed sequence of steps; it does not fan out into a variable number of parallel subtasks. - C is wrong: routing picks one handler for an input; it does not decompose a task or combine results.
Reference: Building effective agents