Free Anthropic CCDV-F Practice Exam with Questions & Answers | Set: 3

Option A is correct because model selection requires measurable non-functional requirements, not an undefined objective such as “best.” Anthropic's model-selection guidance explicitly frames the decision around capabilities, speed, and cost and recommends identifying those requirements before narrowing the candidate models.

“Quality” must also be translated into task-specific success criteria—for example classification accuracy, extraction correctness, acceptable false-positive rate, or human preference. Latency may be constrained by an interactive SLA, and cost may be defined per request, per user, or per completed task. Once those thresholds exist, representative evaluations can compare candidate models using the actual prompts and data.

B uses a generic benchmark that may not correlate with the application's workload. C chooses a model tier without requirements or evidence. D optimizes only for raw capability and ignores latency and cost, potentially producing an unnecessarily expensive architecture.

Thus, requirements definition precedes technology selection. Relevant Study Guide topics: Understanding Reqs, functional versus non-functional requirements, quality thresholds, latency budgets, cost constraints, acceptance criteria, workload-specific evaluation, and model-selection tradeoffs.

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

The supplied examination set identifies A as correct. A small collection of high-quality labeled examples is ideally suited to few-shot or multishot prompting . The examples demonstrate concretely what acceptable input/output behavior looks like, allowing Claude to infer formatting, structure, tone, and task-specific conventions without requiring model retraining.

Anthropic's official prompting guidance states that examples are among the most reliable ways to steer Claude's output format, tone, and structure. It recommends using relevant, diverse examples and clearly separating them from the surrounding instructions. Anthropic currently recommends approximately three to five examples where practical and suggests XML structures such as < examples > and < example > to make prompt organization explicit.

B discards useful supervision by relying exclusively on zero-shot behavior. C is unnecessary for a small fixed set and does not ensure those examples are actually visible to Claude unless additional retrieval logic is created. D introduces unnecessary training complexity for a behavior that prompting already addresses efficiently.

Therefore, A provides the lowest-complexity, highest-leverage solution.

Relevant Claude Developer topics: Agent Construction, multishot prompting, few-shot learning, labeled examples, prompt design, output formatting, behavioral steering, and prompt evaluation .

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

C converts validation failure from an uncontrolled terminal condition into a first-class recoverable error path . The supplied exam item identifies C as correct. If structured output does not meet the application's contract, it should never be forwarded as though valid, but immediate user-visible failure is also unnecessary when bounded recovery is possible.

A robust flow can retry generation, ask Claude to repair the malformed structure using the validation error as feedback, switch to an approved fallback path, or ultimately return a controlled failure if the retry budget is exhausted. The application must cap these recovery attempts to avoid unbounded loops.

Anthropic's Structured Outputs documentation explains that unconstrained model generation can produce parsing errors, missing fields, inconsistent types, or schema violations that otherwise require error handling and retries. Current Structured Outputs can eliminate many schema-level failures through constrained decoding, although exceptional conditions such as refusal or output truncation still require explicit handling.

A violates the validation boundary. B removes a protective control. D transfers an engineering reliability responsibility to end users.

Relevant Claude Developer topics: Claude App Design, structured output, validation, retries, repair loops, fallback logic, bounded recovery, error paths, and resilient downstream integration .

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

The supplied exam source marks B as correct. The first debugging action should be evidence collection rather than immediately modifying prompts or configuration. A trace exposes the complete conditions under which the refusal occurred: user input, system instructions, accumulated conversation state, tool results, parameters, and resulting model output.

This is important because a refusal can originate from several interacting causes. A particular input pattern may trigger it; system instructions may unintentionally conflict with the requested task; prior conversational context may alter Claude's interpretation; or contextual information may make an otherwise acceptable request appear out of scope. Without examining the failing execution path, changing instructions is speculative.

Option A jumps directly to remediation before identifying the cause. C provides useful aggregate analysis but is more appropriate after individual failing executions have been inspected. D is also valuable for comparative debugging, but determining what changed requires the detailed execution data that traces provide.

The correct troubleshooting sequence is therefore: capture or inspect representative failed traces, determine the trigger, compare against successful interactions, formulate a hypothesis, make a targeted change, and re-evaluate.

Relevant Claude Developer topics: Systems Life Cycle, observability, production debugging, tracing, failure analysis, refusals, regression investigation, and evaluation-driven remediation .

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

D applies configuration as code and is the answer explicitly selected in the supplied examination source. Claude Code treats project configuration as part of the repository's working context. Anthropic documents project-level CLAUDE.md as the mechanism for providing project instructions and context, while shared project configuration belongs in .claude/settings.json. Crucially, Anthropic states that .claude/settings.json is intended to be checked into source control and shared with the team .

Keeping these files alongside the code establishes a single versioned relationship between application behavior and the configuration that influences Claude Code. Standard pull requests then provide diffs, peer review, ownership controls, CI validation, audit history, and straightforward rollback. A particular commit can therefore reproduce both source code and its associated Claude project configuration.

A creates competing copies and synchronization risk. B artificially separates behavior configuration from the code version it affects. C creates the same problem specifically for settings.json.

Developer-local values remain distinct: .claude/settings.local.json exists specifically for settings that should not be committed.

Relevant Claude Developer topics: Confia Management, CLAUDE.md, settings.json, configuration as code, source control, pull-request review, reproducibility, auditability, and team-shared Claude Code configuration .

B is correct because the architecture cannot be chosen intelligently until the team defines which quality attribute is more important to the actual business use case . The examination source explicitly selects B. Whole-document and chunked processing each perform better on different dimensions, so there is no universally superior option.

Anthropic's evaluation guidance emphasizes defining concrete success criteria first and validating candidate approaches against representative examples. The correct sequence is therefore to determine whether cross-section reasoning or citation precision is the more critical requirement, establish measurable acceptance thresholds, and evaluate both strategies on representative legal contracts.

Anthropic's legal summarization guidance also confirms that chunking or meta-summarization is useful for long documents, while acknowledging that processing an entire document can be appropriate when it fits within Claude's context window. For citations specifically, Anthropic provides native citation support that chunks document text into citation-addressable units and improves source-grounded citation behavior.

A and C each privilege one quality dimension before establishing business priority. D focuses on cost and latency even though the unresolved problem is quality tradeoff.

Relevant Claude Developer topics: Understanding Reqs, success criteria, requirement prioritization, long-context processing, chunking, citation accuracy, cross-document reasoning, representative evaluations, and architecture tradeoffs .

Option C establishes the strongest application boundary between probabilistic model generation and deterministic downstream processing. When another component requires JSON with a known contract, the application should explicitly define the expected structure and ensure that model output conforms to it before downstream execution. Anthropic’s current Structured Outputs guidance states that structured outputs constrain responses to a specific schema and are intended to provide valid, parseable data for downstream processing. The current Claude API supports JSON Schema through output_config.format, while SDK helpers can additionally parse and validate returned data.

The underlying engineering principle remains the same even when structured-output enforcement is unavailable: never allow unvalidated model-generated structures to become trusted machine input. Option A provides insufficient contractual control. Option B defines the schema but pushes validation too late, increasing the probability that malformed or semantically invalid data reaches dependent components. Option D sacrifices machine reliability entirely.

Therefore, C correctly combines schema specification, explicit format guidance, and validation. This corresponds to Claude Developer topics covering structured outputs, defensive application design, schema validation, and reliable model-to-system interfaces. The question and options are reproduced from the supplied examination set.

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

Option B establishes the correct trust boundary. Fluent or confident language is not evidence that a generated claim is factually correct. If downstream systems treat output as authoritative data, the application must independently establish whether the output meets its correctness requirements before accepting it.

Validation can take several forms depending on the workload: compare generated facts against authoritative records, require citations or source references, constrain output to retrieved evidence, apply deterministic business rules, or use separate evaluation/classification stages. Anthropic's agent engineering guidance repeatedly emphasizes explicit evaluation criteria and validation rather than relying on apparent confidence.

A confuses sampling behavior with factual reliability. Lowering temperature does not establish factual correctness and may only make an incorrect answer more repeatable. C supplies maximum oversight but is unnecessarily expensive and removes useful automation even for low-risk, easily validated cases. D communicates uncertainty to users but does not protect downstream systems that automatically consume the response.

Therefore, B treats model output as untrusted until verified to the level required by the application. Relevant Study Guide topics: output validation, grounding, factuality, confidence calibration, source verification, trust boundaries, and downstream safety.

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