Examworthyexamworthy.comClaude Certified Developer - Foundations cheat sheet
Anthropic
Free to share. Examworthy is not affiliated with or endorsed by Anthropic; CCDV-F and related marks belong to their respective owners.
At a glance
Format: Proctored multiple-choice and multiple-response; each item states how many responses to select
Domain weight map
Heaviest first - spend your time hereHow this exam thinks
This exam rewards the engineer who reads the stated constraint first and lets it eliminate the options, rather than the one who knows the most features.
Spot the trap
Tempting wrong answers, and why they failTempting but wrong
I can instruct the model in the system prompt to state the policy revision it used, and sample answers monthly to confirm it complies.
Why it fails
Tempting because the answers would usually carry a revision label, but an instruction is guidance the model can omit under pressure, so the audit trail would depend on generated text rather than on a recorded fact.
Applications and Integration
Tempting but wrong
A character count is finer grained than a word count, so surely it is a stable proxy for the model's input budget.
Why it fails
It is tempting because characters are finer grained than words, but the ratio of characters to tokens still varies by script and by content, so a fixed character ceiling either rejects valid prose or lets dense text through. Only a tokeniser measures the unit the request is bounded by.
Model Selection and Optimization
Tempting but wrong
If an agent can call the same five tools, giving it the steps in its system prompt and letting it decide the order should be equivalent and save me writing orchestration code.
Why it fails
Tempting because a model with the same five tools looks equivalent and needs less orchestration code. It is wrong because the order becomes a model decision that can differ per invoice, which defeats both the per-step failure attribution and the reproducibility requirement.
Agents and Workflows
Tempting but wrong
Surely I can just let the markup accumulate and rely on compaction to summarise it once the working context nears its limit?
Why it fails
Tempting because compaction does reclaim room in a long conversation, but it is the wrong instrument here: it summarises material that should never have been carried in the first place, and the payloads keep arriving at the same rate.
Prompt and Context Engineering
Tempting but wrong
The query rules are only a few lines, so I can put them in each application's system prompt and circulate an updated prompt whenever the service changes.
Why it fails
Tempting because a system prompt is the fastest place to add behaviour and needs no new infrastructure. It is wrong because it puts the platform team's logic inside four codebases they do not own, so every change to the service becomes four coordinated edits.
Tools and MCPs
Tempting but wrong
If I drop the sampling temperature, the assistant will stick to its system prompt and stop acting on wording it finds inside message bodies.
Why it fails
Tempting because a lower temperature does make output more deterministic, so it feels like a way to make the assistant obey its instructions. It is not a security control: an injected instruction that the model treats as a legitimate request is followed just as reliably at low temperature.
Security and Safety
Tempting but wrong
Conventions should go in settings.json for the project, since that file is committed and read before each request.
Why it fails
settings.json is genuinely committed and reviewable, which makes it tempting. It configures tool behaviour and permissions rather than carrying prose conventions, so the rules would not reach the model as context at all.
Claude Code
Tempting but wrong
The assistant gave a delivery date it could not know, so the tool description must not be firm enough about calling the tracking tool first.
Why it fails
Tempting because tool descriptions do influence tool selection. It is wrong here because the trace shows the tool was called at turn 2, so rewriting the description treats a fault the trace has already ruled out.
Eval, Testing, and Debugging
Key terms
Exam-day rules
- Read the last line first. The stated constraint, whether a latency budget, a nightly volume, an ownership boundary or a measured error rate, is the whole question, and the options only mean anything against it.
- Ask whether a person is waiting. That single question settles the batch-versus-realtime decisions, and it recurs in the largest domain on the exam.
- Prefer the mechanism that enforces over the instruction that requests. When a requirement says never, the answer is the hook, the schema or the removed permission, not the better-worded rule.
- Watch for the option that upgrades the model to fix a problem the model did not cause. A stale fact wants grounding, a truncated answer wants a raised output allowance, and a flaky test wants a different assertion.
- Note how many responses the item asks for and select exactly that many. Multiple-response items state their own count, and a right answer short of the count still scores nothing.
Revision schedule
- Day 1Read the blueprint against something you have built
- Week 1Build the reference application
- Weeks 1-2Go deep on Applications and Integration
- Weeks 2-3Work the optimisation and agent blocks
- Week 3Cover context, tools, MCP and security