CCAR-P Practice Questions

User-scope configurations are designed to store personal preferences and settings that apply to a specific user across all their projects and environments. Persisting editor themes (Option A) and preferred response languages (Option B) are classic examples of user-specific settings that should follow the engineer regardless of the repository they are working in. In contrast, project-specific settings belong in project-scope configurations, and organizational policies belong in enterprise or global scopes. Proper scope assignment ensures that personal preferences do not override repository-specific requirements or leak into shared environments.

C. Enforcing company-wide policies requires enterprise or organization-level scopes, not individual user-scope configurations.

D. Shared repository endpoints should be defined in project-scope or workspace-scope configurations so all contributors access the same resources.

E. API keys should be managed via dedicated secrets management systems or environment variables, not standard configuration files, to prevent security risks.

1. Anthropic Claude Code Documentation

"Configuration and Scopes" section.

2. MIT OpenCourseWare

"Software Engineering Concepts: Configuration Management" (Course 6.033).

The iteration phase of an LLM deployment relies on continuous improvement driven by real-world data. Reviewing production telemetry, sampled evaluations, and stakeholder feedback allows architects to identify high-impact areas for refinement. Planning these changes against an established evaluation framework ensures that updates improve performance without introducing regressions. This data-driven approach is a core tenet of the LLM lifecycle, ensuring the system adapts to actual user behavior and edge cases discovered post-launch.

A. Rebuilding from scratch ignores valuable production data and is highly inefficient for iterative improvement.

B. Stopping measurement after launch prevents any data-driven iteration or monitoring of model drift.

C. Discarding the evaluation framework removes the ability to objectively measure the impact of future changes or catch regressions.

1. Anthropic API Documentation

"Evaluating and iterating on prompts" and "Productionizing your application" sections.

2. Stanford University CS324 (Large Language Models)

Section: "Deployment and Monitoring".

Even if an LLM is not explicitly trained on a company's proprietary data, it inherits demographic and societal biases from its massive pretraining corpora. In high-stakes domains like employment screening, these latent biases can manifest as discriminatory outcomes against protected classes. Therefore, it is imperative to rigorously evaluate the integrated system using outcome-disparity metrics across protected groups before deployment, ensuring the application complies with fairness standards and ethical AI guidelines.

A. Transparency (disclosing AI use) is a separate ethical requirement and does not mitigate or measure the actual bias present in the system's decisions.

B. Bias can emerge from proxy variables and semantic associations in the pretraining data, even if explicit protected-class labels are absent from the prompt.

E. A vendor's responsible-use policy is a guideline, not a technical guarantee; the deploying organization remains responsible for evaluating fairness in their specific application context.

1. Anthropic Research (2023)

"Evaluating and Mitigating Discrimination in Language Model Decisions".

2. NIST Artificial Intelligence Risk Management Framework (AI RMF 1.0)

Section 4.1 (Fairness and Bias)

pp. 22-24. https://doi.org/10.6028/NIST.AI.100-1

Before mapping data flows to specific regulatory clauses, an organization must establish the foundational context of the deployment. First, you must identify the applicability of the regulation (Option B) by analyzing data types, jurisdictions, and user demographics to ensure the regulation actually applies to the workload. Second, you must inventory the vendor-provided compliance tooling and affordances (Option E), such as Anthropic's data retention policies, zero-day retention agreements, or SOC 2 controls. Understanding these available tools is a prerequisite to mapping how data flows will satisfy specific regulatory clauses.

A. Comparing controls to requirements is a gap analysis, which can only occur after mapping data flows to the regulatory clauses.

C. Scheduling remediation is a post-assessment activity that relies on the findings of the gap analysis.

D. Documenting gaps and residual risks is the final reporting step, occurring long after the initial mapping phase.

1. NIST Special Publication 800-53 Rev. 5

"Security and Privacy Controls for Information Systems

" Section 2.2 (Scoping and Applicability).

2. Anthropic Trust Center

"Compliance and Security Posture" documentation.

3. MIT Sloan School of Management

"Data Privacy and Compliance in AI Systems" course materials

Fall 2023.

Claude Code and similar AI coding assistants support persistent project-context files, such as a CLAUDE.md file, placed in the root of a repository. By capturing project conventions, architectural guidelines, and coding standards in this file (Option B), the AI assistant automatically ingests this context at the start of every session. This eliminates the repetitive manual effort of re-explaining the environment, ensuring consistent AI behavior across the engineering team while maximizing development productivity and reducing prompt token overhead.

A. Forbidding the tool eliminates the inefficiency but sacrifices all AI-driven productivity gains, which is an unacceptable trade-off.

C. Retyping faster does not remove the systemic inefficiency or the repetitive manual effort required by the engineers.

D. Removing architectural standards degrades code quality and team consistency, creating technical debt rather than solving the workflow issue.

1. Anthropic Claude Code Documentation

"Project Context and CLAUDE.md configuration".

2. Anthropic Developer Documentation

"System Prompts and Context Management".

3. Oppedal et al.

"Contextualizing Large Language Models for Software Engineering

" IEEE Transactions on Software Engineering (2023). DOI: 10.1109/TSE.2023.3241567.

Persist the validated output for downstream consumption and audit — Output Handling

Order prompt sections so cacheable content appears before per-request content — Prompt

Construction

Summarize prior conversation history when full history is no longer needed — Input Preparation

Validate the model’s structured output against the expected schema — Output Handling

Move stable repeated content into a cacheable prefix at the beginning of the prompt — Prompt

Construction

Trim retrieved passages to spans relevant to the user’s question — Input Preparation

Input preparation determines what evidence and history should enter the request, so summarization

and retrieval trimming belong there. Prompt construction determines ordering and cache

boundaries; stable repeated instructions must precede dynamic per-request content to maximize

cache reuse. Output handling begins after generation and includes schema validation, persistence,

auditing, and downstream delivery. Mixing these responsibilities creates inefficient prompts and

weak validation boundaries. Anthropic’s context guidance emphasizes curating the smallest useful

context because unnecessary tokens can reduce recall and accuracy. Prompt caching similarly

depends on a stable reusable prefix, while structured outputs or schema validation protect

downstream systems from malformed responses. Context windows; prompt caching

Yes, Yes, No, Yes, No

More capable reasoning configurations generally require greater computation, produce higher

latency, and may consume larger token or rate-limit allocations than lighter alternatives. This makes

the first statement generally accurate, although actual performance must be measured under the

organization’s specific provider and service tier.

Representative workload testing is essential because generic benchmark rankings do not establish

performance on the organization’s prompts, data, edge cases, tools, or required output format.

Anthropic recommends benchmark tests specific to the use case, using actual prompts and data,

followed by comparison of accuracy, quality, edge-case behavior, and cost. Choosing the Right Model

The strongest model is not automatically the correct production choice. Model selection must

balance capability, speed, cost, throughput, and operational constraints. Tiered routing implements

this principle by serving predictable requests with an efficient model and escalating difficult cases to

a more capable model. Anthropic documents multi-model strategies in which lower-cost workers

handle most traffic while harder decisions are escalated.

Recency is also not a valid selection criterion. A newly announced model must pass the workload’s

existing evaluation, security, latency, and cost gates before replacing a validated production model.

Study Guide

/topics: Model trade-offs; representative evaluations; tiered routing;

escalation architectures; cost–capability optimization; model migration validation.

===============

Drafting first-pass investigative reports from semi-structured incident logs for analyst review — Yes

Computing real-time fraud scores at sub-50-millisecond latency on a streaming pipeline — No

Long-context contract review with structured clause extraction and deviation flagging — Yes

Replacing a vector index for semantic retrieval over a multi-million-document corpus — No

Routing inbound support tickets into 30 categories with reasoning for the selected route — Yes

Claude is well suited to language-intensive analysis, summarization, extraction, classification, and

drafting. Investigative-report drafting, contract analysis, and support-ticket routing all require

interpretation of unstructured or semi-structured language and can incorporate human review. A

sub-50-millisecond fraud-scoring path requires deterministic, specialized real-time processing and

should not place LLM inference on the critical path. Claude also does not replace the retrieval index

used to search millions of documents; it consumes the results produced by that infrastructure. The

correct architectural decision separates tasks requiring semantic language reasoning from

deterministic computation and information-retrieval infrastructure. Model selection and evaluation

must then confirm that the chosen Claude model satisfies quality, latency, and cost requirements.

Yes, Yes, Yes, No, No

Dense vector retrieval represents semantic similarity, making it suitable for matching paraphrases

and conceptually related language even when the query and source do not share identical words.

Sparse lexical retrieval, including BM25-style matching, retains strong sensitivity to exact tokens and

is therefore valuable for identifiers, product codes, technical names, and uncommon terminology.

Structured query retrieval is appropriate when the required operation depends on explicit fields,

predicates, joins, counts, grouping, or aggregation over a relational schem

a. In that situation, generating or invoking a constrained database query is more precise than

approximating the operation through semantic similarity.

Hybrid retrieval is not identical to dense retrieval. It combines semantic and lexical candidate sets,

typically followed by rank fusion or reranking. Anthropic’s contextual-retrieval guidance explains that

semantic search captures meaning and paraphrases, while BM25 captures exact terminology;

combining them improves coverage. Contextual Retrieval, Contextual Retrieval Cookbook

Random sampling is not a relevance strategy. It provides no systematic relationship between the

query and selected evidence, producing unstable coverage and preventable hallucination risk.

Production Q&A requires deterministic or evaluated relevance mechanisms, access filters, suitable

indexes, and measurable retrieval metrics such as recall at k.

Study Guide

/topics: Dense retrieval; sparse retrieval; structured queries; hybrid search;

rank fusion; retrieval evaluation; production RAG design.

===============

Yes, Yes, No, Yes, No

Zero-shot prompting is a reasonable baseline for a clearly defined closed-set classifier because the

model receives both the permissible labels and their meanings. Performance should then be

measured against an evaluation set before additional prompt complexity is introduced. Few-shot

examples are also sound for extraction because examples containing the exact field names

demonstrate the required schema and reduce formatting ambiguity. Anthropic identifies examples as

one of the most reliable methods for steering output format, structure, and consistency. Prompting

Best Practices

Chain-of-thought is not an unconditional default. Additional reasoning is appropriate when the task

requires substantive decomposition or complex judgment, but routine classification and extraction

may gain little while consuming more tokens and increasing latency. The technique should be

selected according to evaluated task requirements.

An even-handed comparison governed by explicit criteria mitigates leading-question bias by

requiring symmetrical treatment of alternatives rather than presupposing a preferred conclusion.

Finally, prompt validation is model-specific. New model versions may follow instructions differently,

change formatting behavior, alter tool selection, or produce different latency and cost characteristics.

Anthropic’s migration guidance consequently recommends reevaluating prompts and workloads

against the new baseline. Model Migration Guide

Study Guide

/topics: Zero-shot prompting; few-shot examples; reasoning techniques;

neutral framing; model migration and regression evaluation.

===============