CCA-P start here · Guide walkthrough · Domain 6

Domain 6: Stakeholder Communication & Lifecycle Management

Domain 6 is the consultant side of the architect's job: finding out what the business really needs, explaining choices to people who are not engineers, agreeing what “good enough” means, handing the system over, and keeping it healthy after launch. Almost none of it was in CCA-F. Questions describe a meeting or a project moment and ask what the architect should do next; the answer is usually the step that turns a vague wish into something measurable and agreed, at the right point in the project.

14% of the exam5 guide lines · 10 practice questionsGuide v1.0

How each section works: the guide line, the idea in one sentence, the explanation, what to know, a mind map (including when the runner-up wins), a worked example, then two exam-style questions. Try each question before opening the answer.

Other domains: D1 · D2 · D3 · D4 · D5 · D6 · D7

6.1 · Conduct structured discovery and requirement gathering

Guide line: “Conduct structured discovery and requirement gathering”

In one sentence: Before choosing any technology, find out the business goal, who the users are, how the work is done today, what measurable result counts as success, and what constraints, data and risks exist.

Discovery is the first phase of a project: structured conversations and investigation to understand the problem before designing a solution. Requirement gathering is the part of discovery that writes down what the solution must do and within which limits. Stakeholders (the people with an interest in the project: sponsors, users, IT, legal) often arrive with a solution already in mind, such as “we need a chatbot” or “we want agents”. The architect's job is to step back to the problem: what outcome does the business want, for whom, and how will we know it worked?

A structured discovery covers seven things. The goal: the business outcome, such as shorter handling time or fewer errors. The users: who will use it and what they need. The current process: how the work is done today, with its volumes and pain points, which gives a baseline (today's measured performance) to compare against. Success criteria: measurable targets, for example “first reply within one hour for 90% of tickets”. Constraints: budget, latency, compliance, and systems that must be used. Data: does the data the solution needs exist, is it accessible, current and allowed to be used? Risks: what happens when the system is wrong, and who is affected.

Data deserves special attention, because missing or poor data sinks more AI projects than model choice does. A plan to answer questions from “our knowledge base” fails if the knowledge base is out of date or scattered across email. A prototype, a small quick build, is useful once success is defined: it tests the riskiest assumption cheaply. Building before success is defined means nobody can tell whether the prototype worked. Sometimes discovery shows that most of the problem needs no LLM at all.

Goal and users→Current process and baseline→Measurable success criteria→Constraints, data, risks→Prototype the riskiest assumption→Design

Know this

Mind map

When the stem says…PickWhy
A sponsor asks for a technology (“a chatbot”, “agents”) with no defined outcomeDiscovery: goal, users, measurable success criteriaThe request is a solution, not a problem
Stakeholders want different thingsBring them together to agree one measurable outcome firstDesigning for competing goals satisfies no one
Success is defined, but nobody has checked the dataAssess the data, then prototype on a real sampleMissing or stale data is the most common project killer
Goal, data and constraints are clear and signed offMove on to designHere more discovery would only delay the project

Worked example

Situation: A retail chain's operations director asks for “an AI agent for our stores” and wants a demo in two weeks. Different managers describe different hopes: answering staff policy questions, forecasting stock and writing rotas.

Decision: Run a short structured discovery. Pick the use case with the clearest value, measure today's baseline (for example, how long staff wait for policy answers), agree a measurable target, check that the policy documents are current and complete, then prototype that one use case.

Why: The request names a technology, not a problem, and three goals compete. Building a broad demo agent quickly is the tempting move; it would show something, but nobody could say whether it succeeded.

Rule to remember: If success is not yet defined, run discovery and agree a measurable target first; if success is defined but the data is unchecked, assess the data and prototype; if both are settled, move on to design.

Practice

Practice 6.1.1 · Choose one.

A logistics company's COO asks the architect to “put Claude on our customer emails” and start building this week. In the first meeting nobody can say what result would count as success: one manager mentions faster replies, another mentions fewer complaints. What should the architect do first?

  1. Build a quick prototype that drafts replies, and use it to gather opinions.
  2. Agree a measurable success target and today's baseline with stakeholders.
  3. Compare model options on a sample of emails to estimate cost and quality.
  4. Write a full requirements document covering every type of customer email.
Show answer

Answer: B. The deciding fact: nobody can say what success is, and two managers want different things. Until one measurable target and a baseline are agreed, nothing built can be judged.

  • A ❌ Works once success is defined, to test an assumption; now opinions would just reflect the two competing goals.
  • B ✅ Agree a measurable success target and today's baseline with stakeholders.
  • C ❌ Works later in design, but you cannot judge quality before agreeing what quality means for this business.
  • D ❌ Works as thorough documentation, but it skips agreeing the goal and spends weeks on detail that may not matter.
Practice 6.1.2 · Choose two.

A hospital group wants an assistant that answers nurses' questions from its clinical procedures library. The goal and success target are agreed and signed off: answers in under a minute, judged correct by senior nurses 95% of the time. Nobody has yet looked at the library itself. Which two steps should come next?

  1. Choose the most capable model so that the accuracy target is easier to meet.
  2. Write the full system design so that build work can begin straight away.
  3. Check the procedures are current, complete and accessible in one place.
  4. Interview every ward manager again to reconfirm the goal and the target.
  5. Prototype on a sample of real procedures and score it against the target.
Show answer

Answer: C and E. The deciding facts: success is already agreed, but the data has not been examined. The next steps check the library and test the riskiest assumption, that it can support 95% correct answers, with a small prototype.

  • A ❌ Works as a later design choice, but the answers can only be as good as the procedures, which nobody has checked.
  • B ❌ Works once the data is known, but designing now could build on a library that turns out to be stale or incomplete.
  • C ✅ Check the procedures are current, complete and accessible in one place.
  • D ❌ Works if the goal were unclear, but it is already agreed and signed off, so this only delays the project.
  • E ✅ Prototype on a sample of real procedures and score it against the target.

6.2 · Communicate architectural decisions and trade-offs

Guide line: “Communicate architectural decisions and trade-offs”

In one sentence: Explain each decision in the terms the listener cares about, show the real options with their trade-offs and a clear recommendation backed by evidence, and write the decision down.

Architecture decisions are trade-offs: more accuracy usually means more cost or latency, and more automation means more risk. Stakeholders cannot judge a trade-off they cannot see. The architect's job is to make it visible in the listener's language, then recommend.

Tailor to the audience. Executives care about value, cost, risk and time: speak in cost per resolved case, hours saved, error rates and what happens when the system is wrong. Engineers care about components, interfaces and failure handling. Risk and legal teams care about data flows, controls and audit. It is the same decision with different framing. When someone else owns the decision, present two or three real options, each with its cost, quality, latency and risk, then a recommendation with the reason, backed by eval results rather than opinion.

Record decisions in an ADR (architecture decision record): a short document stating the context, the decision, the options considered and the consequences. ADRs stop settled debates from restarting and tell future teams why things are the way they are. Be honest about limits: an LLM system will sometimes be wrong, its outputs vary between runs, and results on real traffic may differ from the test set. Hiding this to win approval costs trust later, at the first visible mistake.

Frame it in the listener's terms→Show the real options and trade-offs→Back them with eval evidence→Recommend one, with the reason→Record it in an ADR

Know this

Mind map

When the stem says…PickWhy
An executive questions the cost of the designCost per correct outcome, with eval numbers and optionsSpeaks to the decision they own
A board or sponsor must decide in a short meetingValue, risk, cost and a recommendation first; detail on requestThey need what the decision depends on
Engineers will build and run itComponent and interface detail, failure handlingThey need the how, not just the why
A decision is likely to be questioned again laterWrite an ADRKeeps the context and reasons after people move on
Only one option meets a hard constraint (data must stay in region)State the constraint and the single optionHere inventing alternatives would waste time

Worked example

Situation: A CFO asks why the support assistant uses a model that costs five times as much per call as the cheapest one available.

Decision: Show the eval results in business terms: the cheaper model resolves 70% of tickets correctly and the chosen one 92%, and each unresolved ticket costs a human handling fee. Present three options (cheaper model only, chosen model only, and routing simple tickets to the cheaper model) with the cost per resolved ticket for each, and recommend routing.

Why: The CFO owns cost, so cost per correct outcome answers the real question. A technical explanation of model capability is the tempting reply; it is accurate, but it does not tell the CFO what the money buys.

Rule to remember: If the listener is an executive, speak in value, cost and risk; if they are engineers, speak in design detail; if someone else owns the decision, give options, trade-offs and your recommendation, then record it in an ADR.

Practice

Practice 6.2.1 · Choose one.

An architect must brief a hospital's board on whether to approve an assistant that drafts discharge letters for clinicians to check. The board will decide in a 15-minute agenda slot. The engineering team has detailed eval results, latency charts and a component diagram. What should the briefing lead with?

  1. The component diagram, so the board understands how the system works.
  2. A live demo of the assistant drafting a letter from a sample record.
  3. Every eval metric the team collected, so that the board has the full picture.
  4. Time saved, error rate with clinician checks, cost, and a recommendation.
Show answer

Answer: D. The deciding facts: the audience is a board making an approval decision in 15 minutes. They need value, risk and cost in their own terms, plus a recommendation; detail can follow if asked.

  • A ❌ Works for engineers who will build it, but a board deciding approval needs value and risk, not components.
  • B ❌ Works to make it concrete, but one demo shows neither the error rate nor the cost the board must weigh.
  • C ❌ Works for a technical review, but in 15 minutes it buries the few numbers the decision depends on.
  • D ✅ Time saved, error rate with clinician checks, cost, and a recommendation.
Practice 6.2.2 · Choose one.

A product director asks the architect how to answer staff policy questions: put every policy into a long context on each request, or use retrieval over a policy index. Both pass the eval; they differ in cost per query and in how quickly policy updates appear. The director says she will make the final call. What should the architect provide?

  1. Both options with their cost and update trade-offs, and a recommendation.
  2. The option the engineers prefer, with the technical reasons for choosing it.
  3. A neutral comparison table of both options, with no recommendation given.
  4. The eval results for both options, so the director can judge their quality.
Show answer

Answer: A. The deciding facts: the director owns the decision, and the options differ on cost and update speed, not quality. She needs both options with those trade-offs, plus the architect's recommendation.

  • A ✅ Both options with their cost and update trade-offs, and a recommendation.
  • B ❌ Works when the decision is delegated to the team; here the director said she decides, so she needs both options.
  • C ❌ Works to show the trade-offs, but the architect is expected to give a view; withholding it leaves the director without expert advice.
  • D ❌ Works as evidence, but both already pass the eval, so quality is not what separates them.

6.3 · Manage stakeholder feedback loops and expectation alignment (including SLAs)

Guide line: “Manage stakeholder feedback loops and expectation alignment (including SLAs)”

In one sentence: Agree early, in numbers, what “good enough” means, including an acceptable error rate and what happens to errors; set measurable SLAs; and keep reviewing with stakeholders as feedback and scope change.

Stakeholders often expect an AI system to be right every time, because demos look flawless. LLM systems are probabilistic: they will make some errors. If that is not discussed early, the first visible mistake looks like failure. Expectation alignment means agreeing, before launch, what quality is acceptable and what happens when the system is wrong.

Two things are agreed together. An acceptable error rate, measured on an eval set that reflects real cases, and usually compared with today's human error rate, which is rarely zero. And error handling: how errors are caught or limited, such as human review of risky cases, easy correction and escalation to a person. An SLA (service level agreement) turns commitments into measurable targets: latency (for example, 95% of responses within three seconds, often written as the p95 latency), availability, and quality thresholds measured on sampled traffic. A target that cannot be measured, like “fast and accurate”, cannot be managed and will be argued about.

After launch, run a feedback loop: collect user feedback and corrections, review the metrics against the SLA with stakeholders on a regular cadence, and turn what you learn into eval cases and improvements. When stakeholders ask for new features, users or countries, treat it as a scope change: assess its effect on cost, timeline and quality, agree a revised plan openly, and update the success criteria, rather than quietly absorbing it or flatly refusing.

Agree acceptable error rate→Agree error handling→Set measurable SLAs→Launch with monitoring→Regular reviews with feedback→Agree scope changes openly

Know this

Mind map

When the stem says…PickWhy
Stakeholders expect 100% accuracyA measured target plus error handling, shown with eval data and the human baselinePerfection is not achievable; a managed error rate is
Sponsor and team disagree whether it is “fast enough”Agree measurable SLAs and track themA vague target cannot settle the argument
A new group or feature is requested mid-projectAssess impact on cost, time and quality, then agree a revised planSilent absorption breaks timelines and quality
Users complain informally and nothing changesA regular review that turns feedback into eval casesCloses the loop
Errors would cause serious harm (medication doses)Human check of every output, even with less automationHere the stricter option wins: error handling must cover every case

Worked example

Situation: A law firm's partners expect a contract-review assistant to catch every risky clause. On 300 past contracts it catches 94% of risky clauses; the firm's junior lawyers, measured on the same set, catch 88%.

Decision: Present both numbers, agree a target (at least 93% of risky clauses caught on a growing test set), keep a lawyer reviewing every contract with the assistant's flags as a starting point, and review misses with the partners each month.

Why: Showing the human baseline turns “not perfect” into “better than today, with a safety net”. Delaying launch until the assistant catches everything is the tempting answer, but no system or person reaches that, so it would never launch.

Rule to remember: If stakeholders expect perfection, agree a measured error target plus error handling; if targets are vague, make them measurable SLAs; if scope grows, assess the impact and agree a revised plan openly.

Practice

Practice 6.3.1 · Choose one.

A telecom sponsor approved a support assistant on the condition that it is “fast and accurate”. Two months after launch, the sponsor says it is too slow, while the engineering team says it meets every target they set themselves. Nobody can show which view is right. What should the architect put in place?

  1. A faster model on the live path, so the sponsor's complaint goes away.
  2. A satisfaction survey sent to customers after each conversation ends.
  3. Latency and quality targets agreed with the sponsor, and tracked openly.
  4. A detailed report from engineering showing the targets they have already met.
Show answer

Answer: C. The deciding fact: the only agreed requirement was “fast and accurate”, so nobody can prove either view. Measurable targets agreed with the sponsor, and tracked where both sides can see them, settle it now and later.

  • A ❌ Works if speed were proven to be the problem, but it guesses, may cost accuracy, and leaves the next dispute unresolved.
  • B ❌ Works as a feedback source, but it measures customer feeling, not the latency and quality the sponsor and team disagree about.
  • C ✅ Latency and quality targets agreed with the sponsor, and tracked openly.
  • D ❌ Works to show the team's view, but those targets were never agreed with the sponsor, so they do not settle anything.
Practice 6.3.2 · Choose one.

Midway through building an HR policy assistant for UK staff, a director asks to add US and German employees before the agreed launch date. Policies, languages and employment rules differ by country, and the agreed success criteria and budget covered UK staff only. What should the architect do?

  1. Add the two countries now, reusing the UK design, to keep the launch date.
  2. Assess the impact on cost, timeline and quality, then agree a revised plan.
  3. Decline the request, since it falls outside the scope agreed at the start.
  4. Note the request in the backlog and pick it up again after the UK launch.
Show answer

Answer: B. The deciding facts: the request changes scope that the criteria and budget did not cover, and the countries differ in rules and language. That is a scope change, so assess its impact and agree the plan with the director.

  • A ❌ Works if the countries were similar, but different rules and languages mean the UK design and evals do not carry over, so quality is at risk.
  • B ✅ Assess the impact on cost, timeline and quality, then agree a revised plan.
  • C ❌ Works to protect the plan, but refusing outright ignores a real business need that a revised plan might meet.
  • D ❌ Works and may be where the assessment ends up, but the architect decides it alone instead of agreeing the impact with the director.

6.4 · Document architectures and provide implementation guidance

Guide line: “Document architectures and provide implementation guidance”

In one sentence: Hand over everything another team needs to run, change and trust the system: the architecture, the decisions, versioned prompts, the eval suite, runbooks, known limitations and named owners.

A Claude system is more than code. Its behaviour depends on prompts, tool descriptions, retrieval settings, model choice and the eval suite (the test cases and scoring that show whether a change helped or hurt). A team that receives only the code repository cannot safely change any of these, because it cannot tell what “working” looks like. Handoff is the point where ownership moves to the team that will run the system, often the client's own team.

A complete handoff pack includes: the architecture (components, data flows, where data is stored and processed); ADRs explaining why each major decision was made; prompts under version control, each version linked to its eval results; the eval suite with its datasets and pass thresholds; runbooks, step-by-step guides for routine operations and incidents (latency spikes, a tool failing, a model version change); known limitations and failure modes; and an ownership table saying who owns each part and who to call.

Implementation guidance for teams building on the design (prompt conventions, how to add a tool, coding standards) belongs in the same pack. Documentation only helps if the receiving team can use it, so test the handoff: walk them through it, have them run the eval suite on a small change and handle a practice incident from a runbook before the engagement ends, then fix whatever they found unclear. Match the pack to what the receiving team says it is missing.

Know this

Mind map

When the stem says…PickWhy
The new team fears changing prompts without knowing the effectThe eval suite tied to versioned promptsProves a change helps, and allows rollback
On-call staff have never handled this system's incidentsRunbooks with clear stepsNobody should rely on memory under pressure
Nobody knows who owns an integrationAn ownership tableIncidents stall without a named owner
The new team asks why a design choice was madeADRsRecord the context and the rejected options
A two-week prototype that will be thrown awayA short summary of findingsHere the light option wins: a full pack would be wasted effort

Worked example

Situation: A consultancy built a claims-triage assistant for an insurer. In six weeks the consultancy leaves, and the insurer's platform team takes over; none of them was involved in the build.

Decision: Prepare the full pack, then run two handover sessions: the platform team runs the eval suite on a small prompt change and handles a simulated tool outage using the runbook, while the consultancy watches and fixes the gaps it sees.

Why: Ownership moves to people with no build context, so they need the reasons and the tests, not just the code. A thorough code walkthrough is the tempting answer; it explains how the code works but not how to tell whether a change broke quality.

Rule to remember: If a team must change the system, give it versioned prompts and the eval suite; if it must run the system, give it runbooks and named owners; if it must understand the system, give it the architecture and ADRs.

Practice

Practice 6.4.1 · Choose one.

A consultancy has built a contract-summarising assistant for a bank, and the bank's own team takes over next month. The bank's team lead says, “We can read the code fine; what worries us is changing a prompt and not knowing whether we've made things worse.” What should the handoff prioritise?

  1. The eval suite and its thresholds, tied to versioned prompts.
  2. A detailed code walkthrough covering every module and dependency.
  3. Runbooks for incidents such as tool outages and latency spikes.
  4. An architecture diagram showing components and data flows end to end.
Show answer

Answer: A. The deciding fact: the team's worry is not knowing whether a prompt change made things worse. The eval suite answers exactly that, and versioned prompts let them compare and roll back.

  • A ✅ The eval suite and its thresholds, tied to versioned prompts.
  • B ❌ Works for teams new to the code, but the lead said reading the code is not the problem.
  • C ❌ Works for operations, and belongs in the pack, but it does not tell them whether a prompt change hurt quality.
  • D ❌ Works to explain the system, but a diagram cannot show whether a change made answers worse.
Practice 6.4.2 · Choose two.

An internal platform team takes over a Claude-based IT helpdesk agent next week and will run it around the clock. Their manager says they already have the code and an architecture diagram, but on-call engineers have never handled this system's incidents, and they keep asking who owns the ticketing integration. Which two handoff items close these gaps?

  1. ADRs explaining the reasons behind each major design decision.
  2. Runbooks with step-by-step actions for the likely incidents.
  3. A versioned prompt library with each version's eval scores.
  4. An ownership table naming who owns each component and integration.
  5. A roadmap of the features planned for the next two quarters.
Show answer

Answer: B and D. The deciding facts are the two gaps named: on-call engineers have never handled incidents, and nobody knows who owns the ticketing integration. Runbooks close the first and an ownership table closes the second.

  • A ❌ Works to explain why the system is built this way, but neither named gap is about design reasons.
  • B ✅ Runbooks with step-by-step actions for the likely incidents.
  • C ❌ Works for teams that will change prompts, but the gaps named are about running it and ownership.
  • D ✅ An ownership table naming who owns each component and integration.
  • E ❌ Works for planning future work, but it does not help anyone handle tonight's incident or find an owner.

6.5 · Support lifecycle phases (discovery, design, handoff, monitoring, iteration)

Guide line: “Support lifecycle phases (discovery, design, handoff, monitoring, iteration)”

In one sentence: An LLM system is never finished: plan for discovery, design, build, handoff, monitoring and iteration, and adopt new model versions through the eval suite and a staged rollout.

Traditional software can stay stable for years once built. An LLM system keeps changing even when nobody touches the code: new model versions arrive and old ones are eventually retired, users ask new kinds of questions, and source documents and data drift (change gradually away from what the system was built and tested on). So the lifecycle is a loop, not a line: discovery, design, build, handoff, monitoring, iteration, and back again.

Each phase has its own output. Discovery gives agreed goals and success criteria. Design gives the architecture and ADRs. Build gives the system plus its eval suite. Handoff gives the documentation pack and trained owners. Monitoring watches quality, cost, latency and safety in production, including sampled quality checks, not only uptime. Iteration turns what monitoring and feedback reveal into changes, each tested on the eval suite before release; production failures become new eval cases, so the suite grows.

Adopting a new model version is a planned change, not an automatic upgrade. Pin the exact model version in production so that changes are deliberate. Run the eval suite on the new version and compare quality, latency and cost with the current one; check that prompts still behave as expected, since a new model can respond differently to the same prompt. Then use a staged rollout: send a small share of traffic to the new version, watch the metrics, and widen step by step, keeping the ability to roll back. Track retirement dates for the versions you use, so a migration is never rushed.

Discovery→Design→Build with evals→Handoff→Monitoring→Iteration, back to design

Know this

Mind map

When the stem says…PickWhy
A newer model version is availableEval suite on it, compare quality and cost, then a staged rolloutNewer is not automatically better for your task
Quality slowly falls with no code or prompt changeLook at data and usage drift, then iterateThe world moved, not the code
The current model version will be retired in some monthsTest the new version now and migrate in stages before the dateAvoids a rushed, untested switch
A production failure was foundFix it and add it to the eval suiteStops the same failure coming back
The new version is better and cheaper on every eval metricStage the rollout quicklyHere moving fast wins, but still in stages

Worked example

Situation: An e-commerce company's product-question assistant runs on a pinned model version. A new version is released, and the product owner, having read that it reasons better, wants to switch tomorrow.

Decision: Run the existing eval suite on the new version, compare accuracy, latency and cost per answer, fix any prompts that behave differently, then send 5% of traffic to it, watch the metrics for a week and widen in steps.

Why: General claims about a new model do not show how it does on this task and these prompts. Switching tomorrow is the tempting answer; it might work, but a regression would reach every customer at once, with no evidence to guide a fix.

Rule to remember: If a new model version arrives, test it on the eval suite and roll it out in stages; if quality drifts with no code change, look at data and usage and add the new cases to the eval suite; if a version is retiring, migrate early.

Practice

Practice 6.5.1 · Choose one.

A support assistant has run for eight months on a pinned model version with no code or prompt changes. Its resolution rate fell from 85% to 76% over the last three months. Sampled conversations show customers increasingly ask about a subscription plan launched in the summer. What should the team do first?

  1. Upgrade to the newest model version, which may handle the questions better.
  2. Roll back the prompt to the version used at launch, when results were better.
  3. Run the existing eval suite again to confirm that it still passes as before.
  4. Add the new plan's content to the sources and its questions to the evals.
Show answer

Answer: D. The deciding facts: nothing in the system changed, and the drop lines up with questions about a new plan. That is usage and data drift, so add the plan's content and cover it in the eval suite.

  • A ❌ Works sometimes, but no model can answer about a plan it has never been given, and the change adds risk.
  • B ❌ Works when a prompt change caused a drop, but the prompt has not changed since launch.
  • C ❌ Works as a check, but the suite predates the new plan, so it will likely pass and hide the problem.
  • D ✅ Add the new plan's content to the sources and its questions to the evals.
Practice 6.5.2 · Choose one.

A bank's document assistant runs on a pinned model version. The provider announces that this version will be retired in four months and recommends a newer one. The bank's change process requires test evidence for any production change. What should the architect plan?

  1. Switch to the newer version on the retirement date, when the change is required.
  2. Switch to the newer version now, since the provider recommends it as better.
  3. Run the eval suite on the new version now, then roll it out in stages.
  4. Ask the provider to extend support for the current version past the date.
Show answer

Answer: C. The deciding facts: four months of notice and a change process that requires evidence. Testing now on the eval suite produces that evidence and leaves time for a staged rollout before the date.

  • A ❌ Works to delay effort, but it forces a rushed switch with no time to fix problems the evals might find.
  • B ❌ Works quickly, but a provider's general recommendation is not the test evidence the bank's process requires.
  • C ✅ Run the eval suite on the new version now, then roll it out in stages.
  • D ❌ Works only to buy time, if the provider agrees, and the migration still has to be tested and planned.

Source of the guide lines: Claude Certified Architect – Professional Exam Guide, v1.0 (Anthropic). Explanations and questions are ours, written fresh. · CCA-P start here