CCAR-F Agentic Architecture and Orchestration Practice Question
When implementing the 'Evaluator-Optimizer' pattern for code generation with Claude 3.5 Sonnet, which TWO components are essential for the feedback loop to be effective?
⚠ Common exam trap
Candidates often forget that an evaluator needs a structured mechanism to communicate errors back, rather than just returning a binary pass/fail status without actionable details.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
A distinct set of evaluation criteria or unit tests for the code
The Evaluator-Optimizer pattern relies on a clear separation of generation and critique. To be effective, the evaluator must have a set of objective criteria or tests to check the generator's work, and the generator must receive the specific results of that evaluation to make targeted improvements in the next iteration of the loop.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
A distinct set of evaluation criteria or unit tests for the code
Why this is correct
Without objective criteria, the evaluator cannot provide consistent or helpful feedback. In code generation, this often takes the form of unit tests or linter rules. These criteria ensure that the optimizer is working toward a measurable goal of quality rather than making arbitrary changes to the code.
- ✗
The use of two different model families to ensure diversity of thought
Why it's wrong here
While using different models (e.g., Sonnet for generation and Opus for evaluation) can be beneficial, it is not an essential requirement. A single high-capability model like Claude 3.5 Sonnet is often capable of both generating code and critiquing it effectively if provided with the right instructions and context.
- ✓
A mechanism to pass the evaluator's feedback back to the generator
Why this is correct
The core of the 'Optimizer' step is the iterative improvement based on feedback. The orchestration layer must capture the evaluator's critique and present it to the generator in a way that the model can understand. This feedback loop is what allows the system to converge on a high-quality final product.
- ✗
Setting the temperature to 0 for both the generator and the evaluator
Why it's wrong here
While low temperature is often preferred for consistency, it is not a functional requirement for the pattern. In fact, slightly higher temperature in the generator can sometimes help the model 'break out' of a suboptimal coding approach when the evaluator identifies a flaw, providing the creative variance needed for problem-solving.
- ✗
A mandatory human approval step after every single model iteration
Why it's wrong here
While HITL is useful, the Evaluator-Optimizer pattern is often designed to be fully automated to iterate quickly. Requiring a human after every step would defeat the purpose of using an LLM-based evaluator to scale the quality control process, making the system much slower and more expensive to operate.
About these practice questions
Courseiva writes every CCAR-F question from scratch — 271 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAR-F practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAR-F exam.