Part 9 — Question Register (Q1–Q150)
All 150 questions are original practice items written for this guide. They are not official Anthropic exam questions and were not taken from any leaked material. The three official sample questions in Exam Guide v1.0 were studied for style only and not reproduced.
Completeness and uniqueness checks
- Numbers Q1–Q150 present exactly once (verified by script): 150 of 150.
- Single-answer items have 4 choices; multiple-response items have 5 choices and state Select TWO, matching the exam guide's statement that each item says how many to select.
- Multi-response items (15): Q13, Q26, Q42, Q49, Q54, Q64, Q72, Q79, Q95, Q103, Q111, Q115, Q127, Q137, Q147.
- Each question tests a distinct scenario; the Scenario column below is the duplicate check (no two rows describe the same situation).
Distribution vs. official blueprint (63-item exam)
| Domain | Official weight | Questions here | Share |
|---|---|---|---|
| D1 | 17% | 26 | 17% |
| D2 | 13% | 19 | 13% |
| D3 | 19% | 29 | 19% |
| D4 | 16% | 24 | 16% |
| D5 | 14% | 21 | 14% |
| D6 | 14% | 21 | 14% |
| D7 | 7% | 10 | 7% |
| Total | 100% | 150 |
Note: the requested 30/35/35/30/20 split was replaced by the official 7-domain weights, as the master prompt allows when the blueprint differs.
Answer-letter balance (single-answer items)
A: 30 · B: 36 · C: 34 · D: 35
Register
| Q | Domain | Difficulty | Answer | Objective | Scenario (uniqueness key) | File |
|---|---|---|---|---|---|---|
| 1 | D1 | Intermediate | B | Translate business problems into Claude-based AI solutions | A retailer's COO asks for "AI to cut returns fraud." The team has years of labeled structu | 09a |
| 2 | D1 | Intermediate | C | Align solutions to business value pillars (efficiency, transformation, | A hospital network has no overnight patient-intake service today. It wants Claude to offer | 09a |
| 3 | D1 | Advanced | A | Translate business problems into Claude-based AI solutions | Legal operations asks to "automate contract review." Attorneys must remain accountable; th | 09a |
| 4 | D1 | Advanced | D | Translate business problems into Claude-based AI solutions | A sponsor wants contract-risk analysis of 100-page documents to be "instant, perfectly acc | 09a |
| 5 | D1 | Intermediate | B | Translate business problems into Claude-based AI solutions | A team built a demo and asks which model to standardize on. There are no acceptance criter | 09a |
| 6 | D1 | Advanced | C | Translate business problems into Claude-based AI solutions | A tax-software vendor must produce identical results for identical inputs; calculations fo | 09a |
| 7 | D1 | Intermediate | A | Select appropriate architectural patterns (workflow, agentic, augmente | A nightly process extracts eight fields from invoices that arrive in three known layouts. | 09a |
| 8 | D1 | Advanced | D | Select appropriate architectural patterns (workflow, agentic, augmente | An SRE team wants help diagnosing production anomalies. Investigation paths are unpredicta | 09a |
| 9 | D1 | Intermediate | B | Select appropriate architectural patterns (workflow, agentic, augmente | A marketing team needs product descriptions that must satisfy a written brand rubric (tone | 09a |
| 10 | D1 | Intermediate | C | Select appropriate architectural patterns (workflow, agentic, augmente | Support tickets fall into five categories; each needs different instructions, tools, and p | 09a |
| 11 | D1 | Intermediate | A | Select appropriate architectural patterns (workflow, agentic, augmente | A reviewer must check a document for four independent risks (privacy, export control, IP, | 09a |
| 12 | D1 | Advanced | D | Select appropriate architectural patterns (workflow, agentic, augmente | A research agent's costs vary wildly; some runs loop for 90 tool calls repeating similar s | 09a |
| 13 | D1 | Intermediate | A, C | Select appropriate architectural patterns (workflow, agentic, augmente | Question (Select TWO): Which are valid reasons to prefer a predefined workflow over an | 09a |
| 14 | D1 | Advanced | B | Design multi-agent systems and orchestration strategies | A single research agent's context fills with raw search results and answer quality degrade | 09a |
| 15 | D1 | Advanced | C | Design multi-agent systems and orchestration strategies | In a multi-agent design, all worker agents share one administrator credential for CRM acce | 09a |
| 16 | D1 | Intermediate | A | Design multi-agent systems and orchestration strategies | A coding assistant must implement a feature that touches an unknown number of files; the n | 09a |
| 17 | D1 | Expert | D | Design multi-agent systems and orchestration strategies | Three parallel worker agents update a shared project plan and sometimes produce conflictin | 09a |
| 18 | D1 | Intermediate | B | Apply decomposition techniques for complex problem solving | Generating a 40-page RFP response in one call yields uneven quality and inconsistent termi | 09a |
| 19 | D1 | Advanced | C | Apply decomposition techniques for complex problem solving | A pipeline was split into 12 tiny prompt steps. Latency tripled; steps don't use different | 09a |
| 20 | D1 | Advanced | A | Apply decomposition techniques for complex problem solving | A single prompt that extracts facts from medical-device complaints and then assesses regul | 09a |
| 21 | D1 | Intermediate | D | Design end-to-end architectures (input → processing → output → feedbac | An assistant has been live for three months. Users can thumbs-down answers, but nothing li | 09a |
| 22 | D1 | Advanced | B | Design end-to-end architectures (input → processing → output → feedbac | A PoC works for five internal users with a hard-coded key, no retries, and no logging. A r | 09a |
| 23 | D1 | Intermediate | C | Design end-to-end architectures (input → processing → output → feedbac | An ADR says "Use a small model for triage," but gives no way to know when to change. | 09a |
| 24 | D1 | Advanced | A | Design end-to-end architectures (input → processing → output → feedbac | Engineering favors an agent; operations favors a workflow; both argue strongly in a design | 09a |
| 25 | D1 | Intermediate | D | Align solutions to business value pillars (efficiency, transformation, | A contact center wants after-call summaries to cut agent after-call work. A baseline of 4. | 09a |
| 26 | D1 | Intermediate | A, D | Select appropriate architectural patterns (workflow, agentic, augmente | Question (Select TWO): Which statements are correct? | 09a |
| 27 | D2 | Intermediate | C | Select appropriate Claude models based on trade-offs | A bank labels 2M customer messages per day into 40 intents with a 500 ms latency budget. O | 09b |
| 28 | D2 | Advanced | A | Select appropriate Claude models based on trade-offs | An assistant answers many simple FAQ questions but occasionally receives complex multi-doc | 09b |
| 29 | D2 | Advanced | D | Select appropriate Claude models based on trade-offs | A newer, cheaper model is released. A manager asks you to switch production traffic immedi | 09b |
| 30 | D2 | Expert | B | Select appropriate Claude models based on trade-offs | A legal-reasoning workflow fails on multi-step cases. You have already clarified the promp | 09b |
| 31 | D2 | Advanced | D | Select appropriate Claude models based on trade-offs | A team concludes "Model X is worse" after running a prompt carefully tuned for Model Y on | 09b |
| 32 | D2 | Intermediate | D | Design system prompts, templates, and guardrails | A summarizer receives user-pasted emails that sometimes contain instructions like "disrega | 09b |
| 33 | D2 | Advanced | C | Design system prompts, templates, and guardrails | A compensation assistant's system prompt says "Never reveal salary data of other employees | 09b |
| 34 | D2 | Intermediate | A | Design system prompts, templates, and guardrails | A downstream parser fails about 3% of the time because the model adds a sentence before th | 09b |
| 35 | D2 | Intermediate | A | Design system prompts, templates, and guardrails | Three teams copy-paste a shared prompt and edit it locally; behavior diverges and no one k | 09b |
| 36 | D2 | Intermediate | D | Apply prompt engineering techniques (zero-shot, few-shot, chain-of-tho | A classifier handles common categories well but mislabels two subtle edge categories that | 09b |
| 37 | D2 | Advanced | C | Apply prompt engineering techniques (zero-shot, few-shot, chain-of-tho | Two features: (1) a latency-critical 6-class routing step; (2) an offline multi-step finan | 09b |
| 38 | D2 | Advanced | B | Apply prompt engineering techniques (zero-shot, few-shot, chain-of-tho | Five few-shot examples add 3,000 tokens to every request and raise cost and latency; quali | 09b |
| 39 | D2 | Advanced | A | Optimize context windows and manage token usage | An agent degrades after about 40 turns. Logs show 90% of tokens are old tool outputs that | 09b |
| 40 | D2 | Intermediate | D | Optimize context windows and manage token usage | Employees query a 300-page policy manual that changes quarterly. | 09b |
| 41 | D2 | Advanced | C | Optimize context windows and manage token usage | Monitoring shows rising cost per request. Tokens are: system 1,200; tool definitions 6,500 | 09b |
| 42 | D2 | Intermediate | A, C | Optimize context windows and manage token usage | Question (Select TWO): Which actions reduce token cost without dropping information th | 09b |
| 43 | D2 | Advanced | D | Implement prompt reuse strategies (caching, modular prompts, Skills) | You added cache_control after a 6,000-token static policy, but usage shows zero cache re |
09b |
| 44 | D2 | Advanced | A | Implement prompt reuse strategies (caching, modular prompts, Skills) | Workload 1: an interactive chat with a long static policy and many users. Workload 2: a ni | 09b |
| 45 | D2 | Intermediate | B | Implement prompt reuse strategies (caching, modular prompts, Skills) | A team has twelve occasionally needed procedures (release notes, incident reports, code re | 09b |
| 46 | D3 | Intermediate | B | Evaluate tool/agent configuration for capability bloat | An internal assistant exposes 45 tools to every conversation. Traces show frequent wrong-t | 09c |
| 47 | D3 | Intermediate | D | Evaluate tool/agent configuration for capability bloat | A business-intelligence assistant only answers read-only questions, but its database tool | 09c |
| 48 | D3 | Advanced | A | Evaluate tool/agent configuration for capability bloat | The tool list includes search_docs, find_documents, and lookup_docs with nearly iden |
09c |
| 49 | D3 | Intermediate | A, C | Evaluate tool/agent configuration for capability bloat | Question (Select TWO): Which observations most strongly indicate capability bloat in a | 09c |
| 50 | D3 | Advanced | C | Analyze authentication and authorization requirements to identify secu | A CRM agent uses a single administrator token for all lookups. A junior rep asks about a r | 09c |
| 51 | D3 | Advanced | B | Analyze authentication and authorization requirements to identify secu | A get_invoices tool takes tenant_id and customer_id as arguments supplied by the mod |
09c |
| 52 | D3 | Expert | D | Analyze authentication and authorization requirements to identify secu | An HTTP-based MCP server accepts any valid OAuth token from the company's identity provide | 09c |
| 53 | D3 | Advanced | A | Analyze authentication and authorization requirements to identify secu | A shared vector index serves several tenants. The app retrieves top-k chunks, sends them t | 09c |
| 54 | D3 | Advanced | A, C | Analyze authentication and authorization requirements to identify secu | Question (Select TWO): Which are authorization gaps? | 09c |
| 55 | D3 | Intermediate | C | Evaluate accuracy-latency trade-offs and justify configuration decisio | A consumer chat feature needs the first words to appear within a second; answers are moder | 09c |
| 56 | D3 | Advanced | B | Evaluate accuracy-latency trade-offs and justify configuration decisio | A second "verification" model call doubles latency. Errors in low-risk draft replies cost | 09c |
| 57 | D3 | Advanced | D | Evaluate accuracy-latency trade-offs and justify configuration decisio | Month-end spikes cause 429 errors. Retries create duplicate refund tickets in the downstre | 09c |
| 58 | D3 | Intermediate | A | Analyze observability challenges and select monitoring strategies at s | Support reports more wrong answers, yet dashboards show 99.9% availability and normal late | 09c |
| 59 | D3 | Advanced | C | Analyze observability challenges and select monitoring strategies at s | Engineers need prompts and tool arguments in logs for debugging, but they may contain pers | 09c |
| 60 | D3 | Advanced | B | Analyze observability challenges and select monitoring strategies at s | At 50M requests/day, storing full traces for all traffic is cost-prohibitive. | 09c |
| 61 | D3 | Intermediate | D | Design a RAG pipeline with appropriate chunking and indexing strategie | Policy documents have headings and tables. Fixed 200-token chunks split tables mid-row, an | 09c |
| 62 | D3 | Advanced | A | Design a RAG pipeline with appropriate chunking and indexing strategie | After one team re-embedded new documents with a different embedding model, retrieval quali | 09c |
| 63 | D3 | Intermediate | B | Design a RAG pipeline with appropriate chunking and indexing strategie | The assistant sometimes cites a superseded benefits policy after a new one was published. | 09c |
| 64 | D3 | Intermediate | A, C | Apply retrieval strategies matched to data shape and query pattern | Question (Select TWO): Users search by exact part numbers like "AX-4410-B" and semanti | 09c |
| 65 | D3 | Intermediate | C | Apply retrieval strategies matched to data shape and query pattern | Analysts ask, "What was Q3 EMEA revenue by product line?" The data lives in a warehouse. | 09c |
| 66 | D3 | Advanced | D | Apply retrieval strategies matched to data shape and query pattern | Questions such as "Which suppliers of components used in Product X had incidents last year | 09c |
| 67 | D3 | Intermediate | A | Evaluate progressive discovery vs. monolithic context strategy | A 40-page, rarely changing product guide is used in almost every support query. | 09c |
| 68 | D3 | Advanced | B | Apply retrieval strategies matched to data shape and query pattern | End-to-end answer quality on a RAG system is 78%. You suspect retrieval but aren't sure. | 09c |
| 69 | D3 | Intermediate | D | Evaluate connection protocols and select the appropriate integration m | Three applications need the same ticketing integration now, and two more teams will add ap | 09c |
| 70 | D3 | Intermediate | C | Evaluate connection protocols and select the appropriate integration m | One internal application calls a single stable internal pricing API on the hot path; laten | 09c |
| 71 | D3 | Advanced | A | Evaluate connection protocols and select the appropriate integration m | A partner exposes an autonomous claims-adjudication agent behind its own API. Your agent s | 09c |
| 72 | D3 | Advanced | B, D | Evaluate connection protocols and select the appropriate integration m | Question (Select TWO): A developer wants to install a popular community MCP server tha | 09c |
| 73 | D3 | Advanced | B | Evaluate progressive discovery vs. monolithic context strategy | An enterprise platform has 200 internal tools; any given task needs about five. Loading al | 09c |
| 74 | D3 | Intermediate | C | Evaluate progressive discovery vs. monolithic context strategy | A focused assistant uses six stable tools that are relevant to almost every conversation. | 09c |
| 75 | D4 | Intermediate | B | Define evaluation metrics (accuracy, latency, cost, safety, security) | A sponsor says the assistant must be "99% accurate." | 09d |
| 76 | D4 | Advanced | D | Define evaluation metrics (accuracy, latency, cost, safety, security) | A candidate release scores 92% on task accuracy (threshold 90%) but yields three critical | 09d |
| 77 | D4 | Advanced | A | Define evaluation metrics (accuracy, latency, cost, safety, security) | The dashboard shows mean cost per request of $0.004, within budget, but finance sees occas | 09d |
| 78 | D4 | Intermediate | C | Define evaluation metrics (accuracy, latency, cost, safety, security) | A legal-summary tool scores well on text-overlap metrics but lawyers say it misses obligat | 09d |
| 79 | D4 | Intermediate | A, C | Define evaluation metrics (accuracy, latency, cost, safety, security) | Question (Select TWO): Which belong in the acceptance criteria for a customer-facing a | 09d |
| 80 | D4 | Intermediate | D | Design evaluation datasets and test frameworks using mixed methodologi | The eval set consists of 100 clean, typical questions. In production, users send typos, of | 09d |
| 81 | D4 | Advanced | B | Design evaluation datasets and test frameworks using mixed methodologi | The team iterates on prompts while reporting results on the same 150-case set, and scores | 09d |
| 82 | D4 | Advanced | A | Design evaluation datasets and test frameworks using mixed methodologi | Overall accuracy is 94%, but accuracy for questions about policies updated this quarter is | 09d |
| 83 | D4 | Intermediate | C | Design evaluation datasets and test frameworks using mixed methodologi | You need to scale the evaluation of open-ended response helpfulness and tone across thousa | 09d |
| 84 | D4 | Advanced | B | Design evaluation datasets and test frameworks using mixed methodologi | An LLM judge consistently prefers longer answers, even when human reviewers prefer concise | 09d |
| 85 | D4 | Intermediate | D | Design evaluation datasets and test frameworks using mixed methodologi | A pipeline extracts invoice totals and dates into JSON, and reference values are known. | 09d |
| 86 | D4 | Advanced | A | Design evaluation datasets and test frameworks using mixed methodologi | A RAG agent with tools gets low end-to-end scores, and the team can't tell where the failu | 09d |
| 87 | D4 | Intermediate | C | Conduct A/B testing and iterative improvements | A new prompt wins on offline evals. Leadership wants to know if it improves resolution rat | 09d |
| 88 | D4 | Advanced | A | Conduct A/B testing and iterative improvements | A cheaper model matches accuracy but fails two indirect-prompt-injection cases that the in | 09d |
| 89 | D4 | Advanced | D | Conduct A/B testing and iterative improvements | A prompt tweak raises average quality by 2 points but drops one customer segment by 6 poin | 09d |
| 90 | D4 | Advanced | A | Conduct A/B testing and iterative improvements | Three days into a planned two-week A/B test, the new variant looks better and a manager wa | 09d |
| 91 | D4 | Intermediate | C | Conduct A/B testing and iterative improvements | An analyst edited the live prompt in a vendor console; quality dropped and no one can tell | 09d |
| 92 | D4 | Intermediate | B | Diagnose system issues (prompt failure, hallucinations, model mismatch | The assistant ignores a length limit that appears at the end of a 3,000-word prompt which | 09d |
| 93 | D4 | Advanced | D | Diagnose system issues (prompt failure, hallucinations, model mismatch | When retrieval returns nothing relevant, the assistant invents a plausible policy. | 09d |
| 94 | D4 | Expert | A | Diagnose system issues (prompt failure, hallucinations, model mismatch | Quality shifted overnight, with no code changes. Your service references a floating model | 09d |
| 95 | D4 | Advanced | A, C | Diagnose system issues (prompt failure, hallucinations, model mismatch | Question (Select TWO): Confident wrong answers began after a nightly index refresh. Wh | 09d |
| 96 | D4 | Advanced | C | Optimize token usage, latency, and cost-performance trade-offs | Cost per request doubled after a prompt edit. Tokens per request rose and cache read token | 09d |
| 97 | D4 | Intermediate | B | Optimize token usage, latency, and cost-performance trade-offs | A workflow calls three independent tools sequentially, adding latency. | 09d |
| 98 | D4 | Advanced | D | Monitor system performance using logging and observability tools | Running an LLM judge over every production output is too expensive. | 09d |
| 99 | D5 | Advanced | D | Implement guardrails and safety controls | An email-triage agent reads inbound customer emails and can send external emails. A penetr | 09e |
| 100 | D5 | Intermediate | B | Implement guardrails and safety controls | An assistant fetches web pages and passes their text to Claude. | 09e |
| 101 | D5 | Intermediate | A | Implement guardrails and safety controls | A public chatbot sees attempts to jailbreak it. You want an inexpensive pre-filter. | 09e |
| 102 | D5 | Advanced | C | Implement guardrails and safety controls | An agent can write files and call external URLs. A review flags exfiltration risk. | 09e |
| 103 | D5 | Intermediate | A, C | Implement guardrails and safety controls | Question (Select TWO): Which are preventive controls? | 09e |
| 104 | D5 | Advanced | D | Implement guardrails and safety controls | A support bot occasionally echoes customers' full card numbers from tool results into repl | 09e |
| 105 | D5 | Intermediate | B | Identify risks, limitations, and failure modes of LLM systems | An assistant has a tool that can delete customer accounts. A prompt injection triggers it. | 09e |
| 106 | D5 | Advanced | A | Identify risks, limitations, and failure modes of LLM systems | A crafted document causes an agent to loop through thousands of tool calls, creating a sur | 09e |
| 107 | D5 | Advanced | C | Identify risks, limitations, and failure modes of LLM systems | Reviewers accept AI recommendations 99% of the time and rarely read evidence. | 09e |
| 108 | D5 | Advanced | D | Apply human-in-the-loop validation strategies | A procurement agent has low confidence on a $40,000 vendor change that cannot be easily re | 09e |
| 109 | D5 | Expert | B | Apply human-in-the-loop validation strategies | A design auto-approves actions when Claude states "confidence: 95%." | 09e |
| 110 | D5 | Intermediate | A | Apply human-in-the-loop validation strategies | The safety-screening service times out during a traffic spike. | 09e |
| 111 | D5 | Advanced | A, C | Apply human-in-the-loop validation strategies | Question (Select TWO): Which changes best reduce rubber-stamp approvals? | 09e |
| 112 | D5 | Advanced | C | Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP) | A payer wants an assistant to summarize member calls containing PHI. Design freeze is next | 09e |
| 113 | D5 | Advanced | D | Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP) | A user in the EU requests erasure of their personal data. The system stores it in a CRM, a | 09e |
| 114 | D5 | Advanced | B | Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP) | A federal agency requires services authorized at a specific FedRAMP impact level. | 09e |
| 115 | D5 | Intermediate | A, C | Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP) | Question (Select TWO): Which implement the principle of data minimization? | 09e |
| 116 | D5 | Intermediate | A | Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP) | An auditor asks you to demonstrate that access controls work. | 09e |
| 117 | D5 | Advanced | C | Address ethical AI considerations (bias, fairness, transparency) | A résumé-screening assistant ranks identical resumes differently when applicant names chan | 09e |
| 118 | D5 | Intermediate | D | Address ethical AI considerations (bias, fairness, transparency) | A benefits assistant influences eligibility guidance. | 09e |
| 119 | D5 | Advanced | B | Address ethical AI considerations (bias, fairness, transparency) | Overall accuracy is 95%, but it is 82% for one language group; the acceptance policy requi | 09e |
| 120 | D6 | Intermediate | B | Elicit and clarify business requirements through discovery | A VP says, "We need an AI agent for customer support." No metrics, volumes, or constraints | 09f |
| 121 | D6 | Advanced | C | Elicit and clarify business requirements through discovery | Two executives give conflicting goals: Operations wants minimum cost per ticket; Legal wan | 09f |
| 122 | D6 | Intermediate | D | Communicate trade-offs and set realistic expectations | A stakeholder asks for "100% accuracy" on policy answers. | 09f |
| 123 | D6 | Advanced | B | Communicate trade-offs and set realistic expectations | Finance asks why the pilot's cost per resolved case is above projections. Analysis shows l | 09f |
| 124 | D6 | Intermediate | C | Build a business case and communicate value | You must write an executive summary for a steering committee. | 09f |
| 125 | D6 | Advanced | D | Build a business case and communicate value | A business case claims "$3M savings" based on automating 100% of tickets. | 09f |
| 126 | D6 | Advanced | B | Define SLAs and manage approvals/objections | The business asks for "99.99% availability and 1-second responses" for an agent that makes | 09f |
| 127 | D6 | Advanced | A, C | Define SLAs and manage approvals/objections | Question (Select TWO): A security team objects that the agent could leak data. Which r | 09f |
| 128 | D6 | Intermediate | B | Define SLAs and manage approvals/objections | A change to the system prompt could alter customer-facing tone. Compliance requires approv | 09f |
| 129 | D6 | Intermediate | C | Manage the solution lifecycle from PoC to production | A PoC with ten curated examples impressed the sponsor, who wants to go live next week. | 09f |
| 130 | D6 | Advanced | D | Manage the solution lifecycle from PoC to production | A pilot serves 5% of traffic. Quality is on target, but cost per case is rising with usage | 09f |
| 131 | D6 | Advanced | C | Manage the solution lifecycle from PoC to production | The provider announces that the model version you use will be deprecated in 90 days. | 09f |
| 132 | D6 | Intermediate | B | Produce documentation and handoff artifacts | The client's team will operate the solution after launch. | 09f |
| 133 | D6 | Advanced | C | Produce documentation and handoff artifacts | Six months later, nobody remembers why a prompt uses a particular retrieval cutoff. | 09f |
| 134 | D6 | Intermediate | C | Produce documentation and handoff artifacts | A production-readiness review is scheduled. | 09f |
| 135 | D6 | Advanced | B | Elicit and clarify business requirements through discovery | Discovery reveals the target process is deterministic, with stable rules and structured in | 09f |
| 136 | D6 | Intermediate | D | Communicate trade-offs and set realistic expectations | The sponsor wants to launch broadly after an impressive pilot, but metrics show weak perfo | 09f |
| 137 | D6 | Advanced | A, C | Manage the solution lifecycle from PoC to production | Question (Select TWO): Which are appropriate post-launch lifecycle practices? | 09f |
| 138 | D6 | Advanced | B | Define SLAs and manage approvals/objections | A customer's contract requires that a human review all legal-risk responses within 4 hours | 09f |
| 139 | D6 | Intermediate | C | Build a business case and communicate value | After launch, the sponsor asks, "Is it working?" | 09f |
| 140 | D6 | Expert | D | Produce documentation and handoff artifacts | A regulated client requires evidence that changes were reviewed, tested, and approved befo | 09f |
| 141 | D7 | Intermediate | A | Configure shared team tooling and standards | A 40-engineer team uses Claude Code. Each developer re-explains build commands, convention | 09g |
| 142 | D7 | Advanced | C | Configure shared team tooling and standards | Security wants to prevent the coding agent from reading secrets files or running destructi | 09g |
| 143 | D7 | Intermediate | C | Establish developer workflows with AI assistance | A developer asks the agent to refactor a large module in one request. The result is a huge | 09g |
| 144 | D7 | Advanced | D | Establish developer workflows with AI assistance | The team wants an agent to review pull requests in CI. | 09g |
| 145 | D7 | Intermediate | B | Create operational runbooks and support processes | At 2 a.m. the on-call engineer sees a spike in refusals and escalations from the productio | 09g |
| 146 | D7 | Advanced | A | Create operational runbooks and support processes | The assistant's behavior changed after a "minor" prompt tweak, and nobody can tell which v | 09g |
| 147 | D7 | Intermediate | A, C | Train and onboard teams | Question (Select TWO): Which practices best onboard a team to a new AI-assisted workfl | 09g |
| 148 | D7 | Advanced | B | Configure shared team tooling and standards | Teams want to reuse a standard code-review procedure with reference checklists and scripts | 09g |
| 149 | D7 | Intermediate | A | Establish developer workflows with AI assistance | A team wants a deterministic action every time the agent edits a file, such as running the | 09g |
| 150 | D7 | Advanced | B | Create operational runbooks and support processes | Leadership asks how to know whether the AI-assisted workflow improves engineering outcomes | 09g |