Pass the Claude Developer Foundations exam
A primer for absolute beginners, then eight guides — one per exam domain, covering every subdomain of the official exam guide — the concepts taught plainly, the numbers the exam quotes, and the traps it sets. No pass guarantee; just nothing missing. About 137 minutes of reading, and the credential lasts a year.
The exam at a glance
Claude Certified Developer — Foundations (CCDV-F) is the fair one of Anthropic’s three certifications — fair because it tests what a working developer who builds with the API actually knows: scenario-based and conceptual, answerable from the public documentation. We sat all three; this is the one we’d point most developers at first.
If you don’t pass: wait 14 days before your second attempt, 30 days before your third, 90 days before your fourth — up to four attempts per rolling year. The full booking timeline and free practice-question sets are in the prep-plan post; these guides are the study material that post promised.
The eight domains
Study where the marks are. Applications and integration (33.1%), Model selection and optimization (16.8%) and Agents and workflows (14.7%) — the three highest-weighted domains — carry 64.6% of the exam between them, so read them first. Claude Code and evals together are 5.7%: a glance, not a study session. Each guide page lists the modules it covers, in exam-guide order.
What Claude actually is, what an AI engineer does all day, and your first API call on paper — the floor the eight guides build on. Five minutes; skip it if you have ever called the API.
| Domain | Weight | Modules | Read | Study priority |
|---|---|---|---|---|
| 01Agents and workflows | 14.7% | 3 | 17 min | 3rd of 8 |
| 02Applications and integration | 33.1% | 6 | 18 min | 1st of 8 |
| 03Claude Code | 3.1% | 1 | 15 min | 7th of 8 |
| 04Eval, testing, and debugging | 2.6% | 1 | 6 min | 8th of 8 |
| 05Model selection and optimization | 16.8% | 4 | 8 min | 2nd of 8 |
| 06Prompt and context engineering | 11% | 3 | 16 min | 4th of 8 |
| 07Security and safety | 8.1% | 4 | 25 min | 6th of 8 |
| 08Tools and MCPs | 10.6% | 3 | 22 min | 5th of 8 |
Who decides the next step — your code or the model — plus the five workflow patterns and the loop an agent actually runs.
The heart of the exam, for people who actually build with the API: request shapes, conversation state, model lifecycles, streaming and the files around a deployment.
The smallest slice, and a dense one: which config file outranks which, plus skills, subagents and unattended headless runs.
Glance-sized by the official weighting — but it is the error-handling kit you will use on every other domain.
The most numeric domain: model tiers, prices, effort levels, thinking budgets, token counting and cost arithmetic.
Two halves: managing what fills the context window, and getting reliable shapes out of the model.
Scenario questions where the plausible-sounding answer is usually the wrong one — one organising principle carries the domain.
How Claude acts: tool definitions that get called correctly, what an MCP server actually is, and choosing the smallest mechanism that closes the gap.
How to use this guide
Read the domains in weight order — that is what the study-order toggle above is for — because hours spent should follow marks available. Each page stands alone: the concepts taught from the ground up, the specific numbers the exam quotes, and a traps section collecting the wrong answers that sound sensible.
Prerequisites: effectively none. If Claude itself — or APIs in general — is new to you, start with the Start here primer: what Claude is, what an AI engineer does, your first API call on paper. If you have ever called an HTTP API, skip straight to the domains — the guides teach from the ground up — though Applications and integration still moves at a working developer’s pace, so budget a second pass through that one.
How this was written and checked: by Fabs, who sat all three Anthropic certifications in 2026 — Developer, Architect Foundations, Architect Professional — and wrote up each one. The structure follows the official exam guide’s eight domains and every subdomain in them; every fact is anchored in Anthropic’s documentation as it stood in September 2026; and coverage was cross-checked lesson by lesson against CertSafari’s 25-lesson syllabus, an independent course. We are not affiliated with or endorsed by Anthropic.
Model names, prices and policies move — if you are reading this in six months, re-check anything numeric against the docs before you memorise it.
One honest caveat, the same one we attach to the certification itself: a certificate proves you studied, not that you can ship. But the studying is a genuinely good tour of the Claude API — and if you pass, you get a year of arguing about agents with a badge behind your name.
Studying for the architect exams instead? The CCAR-F guide covers the foundations certification the same way — five domains, every subdomain, the numbers, the traps — and the CCAR-P guide takes the professional level: seven domains, from solution design to stakeholder lifecycle. All three live under Study guides.
Where to verify us
Anthropic’s own material is free and it is the source of truth. If you have the hours, start there; this guide is the compression — same syllabus, about 137 minutes, the numbers in one place, the traps marked. Every claim on this page is checkable against the links below.
Registration, the exam guide (the source of truth for domains and weights), and policies.
The free official course; most of its hours land on the Applications and integration domain.
Where every fact in this guide is ultimately anchored. Read it like it’s 2003 and you have an exam tomorrow.
A third-party 25-lesson course keyed to the exam guide’s subdomains, plus quiz mode. We cross-checked our coverage against it; useful as a second pass.