CCAR-P Advanced Agentic Architecture Practice Question
An agentic system often experiences 'goal drift' when managing long-running, multi-step tasks. Which architectural pattern most effectively mitigates this risk during recursive reasoning chains?
⚠ Common exam trap
Candidates often suggest increasing the model's context window or using a more powerful model, ignoring that architectural patterns like reflection loops are required to fix logical drifting in multi-step chains.
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
✓
Integrating a reflection loop that evaluates progress against the initial goal.
State-space re-grounding via periodic reflection loops allows the agent to compare current progress against original user intent. By forcing a dedicated 'evaluator' step that analyzes the chain of thought against the task definition, the system can self-correct before executing irreversible actions. This prevents the agent from spiraling into irrelevant sub-tasks that deviate from the primary objective, ensuring high task fidelity in complex autonomous workflows.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Increasing the context window size to include all previous turns.
Why it's wrong here
Expanding context window size does not prevent drift because agents often become biased toward the most recent tokens in long histories. Without a structured reflection mechanism, the model may continue hallucinating or wandering off-task even with total access to previous steps, leading to wasted compute and inaccurate outcomes.
- ✗
Implementing a hard-coded decision tree for every possible action.
Why it's wrong here
Hard-coding logic negates the benefits of agentic autonomy by creating a brittle system that cannot handle edge cases or dynamic environments. Agentic architectures are designed to be flexible and adaptive; static trees lead to high maintenance costs and failures whenever the real-world input deviates from predefined paths.
- ✓
Integrating a reflection loop that evaluates progress against the initial goal.
Why this is correct
Reflection loops force the model to pause and assess the current state against its target objective. This meta-cognitive step allows the agent to identify deviations, prune irrelevant reasoning chains, and re-orient its strategy. It is essential for long-horizon task completion where errors naturally accumulate without periodic corrective oversight.
- ✗
Reducing the temperature parameter to zero for all model calls.
Why it's wrong here
Setting temperature to zero increases determinism but does not prevent conceptual drift. If the underlying reasoning process is flawed, a deterministic model will simply reach the same incorrect conclusion repeatedly. Correctness requires architectural safeguards that validate logic, rather than just reducing the randomness of the model's output generation.
About these practice questions
This CCAR-P question is part of Courseiva's 262-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam 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-P 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-P exam.