Panaversity Certified Developer: Foundations
Maps to CCDV-F, Claude Certified Developer: Foundations. This exam is built against that blueprint and scored to the same standard, so clearing it is the cheapest honest read on whether you are ready to spend a real Anthropic attempt. Read the official CCDV-F exam guide.
This is the exam for the person who ships. You call the API, you wire the tool, you handle the error the model made at three in the morning, and you own the result afterwards.
PCDV-F examines that work. It is Panaversity's developer credential, aligned to Anthropic's Claude Certified Developer – Foundations blueprint, and it asks whether you can design, integrate, and operate an LLM application that survives production rather than a demo.
What it proves to a stranger is narrower and harder than fluency with AI. It proves you can build and ship, not only design. That is why it sits beside the FDE path rather than inside it: architecture judgment says what should be built, and this credential says the thing got built.
Anthropic's guide names the audience for the twin exactly: technical professionals who build, integrate, and ship production-grade AI solutions using large language models, particularly AI and machine learning engineers, technical leads, and senior software engineers working where business requirements meet implementation.
It then says who it is not for, and the list is worth reading before you book. Non-technical or casual users of Claude-based applications. Anyone without hands-on software development experience. Roles limited to prompt writing without wider application development responsibility.
If you are in that second list today, PCAO-F is the exam that fits you, and it is the first step on the FDE path anyway.
The Longer Version of the Same Exam
PCDV-F is the longer version of CCDV-F. Not a practice run at it, not an imitation of it. It asks every objective on Anthropic's Developer Foundations blueprint, at the published weights, and then keeps going past the point where the Anthropic exam stops.
| Panaversity exam | Code | Anthropic twin | Code | Cut score |
|---|---|---|---|---|
| Panaversity Certified Developer: Foundations | PCDV-F | Claude Certified Developer – Foundations | CCDV-F | 720 out of 1,000, both |
The twin is a 53-item, 120-minute, $125 proctored exam. It examines implementation judgment inside one vendor's product surface: the Messages API, the Claude Agent SDK, CLAUDE.md, settings.json, Claude Code.
The Panaversity sitting asks all of that and then asks the professional version of each question. The API contract, not one vendor's Messages API. The agent loop, not one framework's harness. The evaluation discipline, not one product's debugging view. The secrets and identity obligation, not one platform's settings file.
That difference is the whole reason the longer exam exists. You will stand in front of a customer who may run Claude, or may run an open-weights model on their own hardware, or may be six months into a migration between the two. You have to be right either way. An exam that only ever tests one vendor's surface cannot tell you whether you are.
CCDV-F is the credential that travels. It is Anthropic's, an employer can verify it, and it is the one that belongs on your CV. PCDV-F is not a rival to it and does not replace it. It covers the same blueprint before you sit the real one, and adds the portable half of the work.
Panaversity administers and proctors its own exams, holds them to the same 720 out of 1,000 standard Anthropic uses, and runs four of them, each aligned to one Anthropic credential. This is the developer one.
What It Tests
There are eight published domains. These are their weights, taken from the official exam guide.
| Domain | Weight |
|---|---|
| Applications and Integration | 33.1% |
| Model Selection and Optimization | 16.8% |
| Agents and Workflows | 14.7% |
| Prompt and Context Engineering | 11.0% |
| Tools and MCPs | 10.6% |
| Security and Safety | 8.1% |
| Claude Code | 3.1% |
| Eval, Testing, and Debugging | 2.6% |
The weights are the most useful thing on this page, because they are so uneven. Applications and Integration alone is a third of the exam, more than twice the next domain. Add Model Selection and Optimization and you are at almost half the paper before you reach agents, tools, or security.
Look inside the largest domain and the picture sharpens further. The guide breaks it into six objectives and publishes a weight for each.
| Objective inside Applications and Integration | Weight |
|---|---|
| Claude Application Design | 8.6% |
| Software Engineering Foundations | 7.4% |
| Claude API Mechanics | 6.8% |
| Configuration Management | 4.1% |
| Understanding Requirements | 3.4% |
| Systems Life Cycle | 2.8% |
The top three rows there are also the three largest single objectives anywhere on the exam, ahead of every objective in every other domain. This is a software engineering exam with an LLM in it, not an LLM exam with some code attached. REST APIs, JSON, asynchronous programming, version control, code review, and refactoring are examined on their own terms, next to how Claude interprets instructions across interfaces and how CLAUDE.md, settings.json, model pinning, and prompt versioning are managed.
The other surprise is at the bottom. Claude Code and Eval, Testing, and Debugging together are under 6 percent, which is far less weight than either topic usually carries in conversation. Study both. Do not give either the time you give the top of the table.
The Panaversity sitting keeps those eight domains and their weights, then adds the coverage a developer needs in front of a customer who has not standardised on anything: portable API and SDK patterns across providers, evaluation and regression discipline that is not tied to one product's tooling, cost and latency reasoning when the model behind the endpoint can be swapped, and the security and identity obligations you carry whoever hosts the weights.
What to Read
The book is strongest on some of the small domains and thinner on the largest one, so read the top row hardest.
| Domain | Read |
|---|---|
| Applications and Integration | Python in the AI Era, Deploy Your Agent Harness, Claude Code and OpenCode |
| Model Selection and Optimization | The Four Layers, Open Source LLMs |
| Agents and Workflows | Build AI Agents with the OpenAI Agents SDK, Loop Engineering, Choosing Agentic Architectures |
| Prompt and Context Engineering | AI Prompting in 2026, Building the Context Layer, RAG on Postgres |
| Tools and MCPs | Connector-Native Apps, Skills & Connectors, Give Your AI Agent a Nervous System |
| Security and Safety | AI Identity |
| Claude Code | Claude Code and OpenCode |
| Eval, Testing, and Debugging | Trusting the Checker, Eval-Driven Development |
Two further courses pay for themselves inside that 33.1% row. The Loop by Hand is where the agentic loop, stop_reason, and returning tool results correctly are taught end to end. Structured Extraction Pipelines covers JSON schemas, structured output, validation and retries, and batching, which is the shape most application-and-integration questions take.
Two areas move faster than any written course can track, and you should plan to read current documentation for them rather than expect a chapter. Claude model-selection economics: the present Opus, Sonnet, and Haiku trade-offs across quality, latency, and price, plus token budgeting, prompt caching, and batch economics. The rest of the Messages API surface: vision and multimodal input, extended thinking, prompt-caching mechanics, and provider-specific behaviour on Bedrock, Vertex, and Foundry.
Sit a Sample First
The Panaversity PCDV-F sample exam is coming soon. It is not published yet, and there is no date to give you. When it ships it will be free and it will not count against your attempts.
Until then, two things are worth your time. The free Anthropic Academy courses are Anthropic's own material and cost nothing: for this exam, Building with the Claude API, Introduction to MCP, Claude Code in Action, Introduction to agent skills, and Introduction to subagents. And Section 8 of the official CCDV-F guide carries three illustrative items with a worked answer key, though the guide is explicit that they are not drawn from the live item bank.
Use both to find weak domains, and treat the official exam guide as the authority on what is tested.
Schedule Your Exam
PCDV-F is proctored. There are two routes to a seat, and the card below opens the request form for whichever one is yours.
Two attempts are free for students enrolled in the matching Panaversity course. A third or later attempt is charged at a fee still to be announced, and candidates who are not enrolled pay a proctoring fee that is also still to be announced.
After You Pass
Be clear about what this credential does and does not open.
PCDV-F is not part of the FDE gate. That gate is PCAO-F first, then PCAR-F. Passing both is what qualifies a candidate for the Panaversity FDE Internship Program and the partner access that carries the organisational account needed to register for Anthropic's exams. Passing PCDV-F alone does not.
What it does is prove the build. For a Vertical FDE it is the credential that answers the question an architecture credential leaves open: whether the person who designed the system can also stand it up, integrate it, instrument it, and keep it running. For a senior software engineer moving into agentic work, it is the cleanest independent statement that the move has actually happened.
One structural detail is worth knowing before you choose your order. On the Anthropic side, CCDV-F counts toward Claude Partner Network tier eligibility, alongside Architect Foundations and Architect Professional. The Associate credential does not. If your employer's partner standing matters to you, the developer exam is not an optional extra.
Take it after PCAO-F and PCAR-F if you are on the FDE path, or take it first if your work is building and you want the proof that matches it.
The Anthropic Twin
| Field | Value |
|---|---|
| Full name | Claude Certified Developer – Foundations |
| Exam code | CCDV-F |
| Fee | $125 USD, before any partner-tier discount applied at checkout |
| Items | 53, multiple-choice and multiple-response; each item states how many responses to select |
| Time limit | 120 minutes |
| Delivery | Proctored, online or at a test centre, through Pearson VUE |
| Passing score | Scaled score of 720 on a scale of 100 to 1,000 |
| Result reporting | Pass or fail with the scaled score, plus percent-correct by domain |
| Validity | 12 months from the date the credential is awarded |
| Guide version | Version 1.0, effective July 2026 |
720 is a scaled score on a 100 to 1,000 range, set by a standard-setting study. It is not 72 percent correct, and the guide never publishes a percent-correct passing threshold. The per-domain percentages on your score report are informational and are not used to decide pass or fail.
Recertification. The credential is valid for 12 months. To renew on time you review what has changed since you certified and complete a free, non-proctored assessment on the Anthropic Partner Academy. There is no fee for on-time renewal. Let it lapse and you retake the full exam at the full fee. If exam content changes significantly, Anthropic may require a full retake instead of the renewal assessment.
What a failed attempt costs. The fee applies to every attempt. After a first failure you wait 14 days, after a second 30 days, after a third 90 days, and you may make at most four attempts at the same exam in a rolling twelve months. You can cancel or reschedule free up to 24 hours before the appointment; inside that window, and for a no-show or a late arrival, the fee is forfeited. You accept a confidentiality agreement before the exam begins, and declining it ends the session with no refund. Appeals go to Pearson VUE within 14 days, and neither the standard-setting outcome nor individual item content is appealable. This is the arithmetic a free Panaversity attempt exists to spare you.
Prerequisites. There are none. The guide states plainly that no course or prior certification is required, and that the recommended experience, roughly one to five years of software engineering with six months or more hands-on with Claude or comparable LLM systems, Python or TypeScript, and fluency with REST APIs and CLI tools, is recommended rather than required. It also says there is no single required course and that no resource guarantees a passing result, which applies to this book as much as to anything else.
The sources. Download the Claude Certified Developer, Foundations, Exam Guide v1.0 before you book, and register through the Anthropic Partner Academy. Anthropic's Exam Registration Guide walks the process from Partner Academy registration to Pearson VUE scheduling.
The direct PDF links can change. The guide header warns that it is subject to change without notice. If a link stops working, open the exam's Partner Academy page and download the current guide there. The Partner Academy is the durable source.
Where to Go Next
The certification path is the overview: both stages, what each exam costs, and the order to take them in.
The other three exams are PCAO-F: Associate, Foundations, which is the required first step on the FDE path, PCAR-F: Architect, Foundations, which is the second and completes the internship gate, and PCAR-P: Architect, Professional, the capstone for architects who own a solution end to end.