CCDV-F reports on a scaled range of 100–1000 with a cut at 720, and its questions are distributed across 8 domains by published weight rather than evenly. Those weights are the single most useful thing in the exam guide: they tell you where the marks are, and they rarely match where candidates spend their time.

Applications and Integration is the heaviest at 33.1% — roughly 18 of the 53 questions. Weakness there costs more than weakness anywhere else, which is exactly what a flat percentage score hides and what a blueprint-weighted estimate surfaces.

Where to start

Three objectives carry the most weight on CCDV-F. Between them they are worth roughly 13 of the 53 questions:

Which is not the same list as the objectives that feel hardest, and that gap is where study time usually goes missing. A week spent on the topic that worries you most is a week spent on 9% of the exam at best, and often on far less.

How we practice it

Our CCDV-F bank holds 406 original items across 24 objectives, each with an explanation and a note on every option — why the right answer is right, and why each distractor fails. Items are written against the published objectives. None is recalled or reproduced from a live exam.

A drill allocates questions by the published domain weights, multiplied by how far your measured mastery sits below standard. So a domain worth 33% of the exam that you are weak in dominates your next sitting, while one you have already mastered drops back to an exposure floor that keeps its estimate current.

What a question looks like

Taken from the free CCDV-F set, not written for this page — the correct answer, why it is correct, and a note on why each other option fails. Scenarios, not recall: nothing here is answerable from a definition.

CCDV-F Agent Construction with Claude

Using Managed Agents, where do the model, system prompt, and tool definitions belong?

A On each session, so that every run can be configured independently of others

Putting configuration on sessions defeats the versioning the surface exists to provide, and it makes every run a fresh definition.

B On the persisted agent object, which sessions then reference by its stored ID

Correct. Configuration lives on the persisted, versioned agent, and each session pins to it by ID — create the agent once, start a session per run.

C In the first user message of each session, where the model reads them directly

Instructions in a user message are conversation content, not configuration, and they would have to be repeated on every run.

D In the environment definition, alongside the container and networking settings

The environment describes the container and its networking; it does not carry the model or the prompt.

The mandatory flow is agent once, session per run: configuration lives on the versioned agent object and sessions pin to it. Putting configuration on sessions defeats the persistence and versioning the surface exists to provide.

What you get for free

A 20-item diagnostic for CCDV-F, spread across objectives rather than taken off the top, so the sample reflects the shape of the exam. It produces a first readiness estimate and full explanations, whether or not you go on to buy.

The other Claude certification exams

All four report on the same 100–1000 scale with the same 720 cut, so a readiness estimate means the same thing whichever you are sitting. What differs is shape:

Domain	Share of exam	Approx. questions
Domain 1 — Agents and Workflows	14.7%	8
Domain 2 — Applications and Integration	33.1%	18
Domain 3 — Claude Code	3.1%	2
Domain 4 — Eval, Testing, and Debugging	2.6%	1
Domain 5 — Model Selection and Optimization	16.8%	9
Domain 6 — Prompt and Context Engineering	11.0%	6
Domain 7 — Security and Safety	8.1%	4
Domain 8 — Tools and MCPs	10.6%	6

Claude Certified Developer — Foundations

What the exam covers

Where to start

How we practice it

What a question looks like

What you get for free

The other Claude certification exams