30-Day Structured Plan (CCAO-F)¶
Level: Beginner · Time: 1 to 1.5 hours a day, ~38 hours total · Track: CCAO-F (Claude Certified Associate – Foundations)
Who this plan is for¶
This plan is written for complete beginners. You do not need to have used an AI model before, and you do not need to write code. If you can use a web browser and set aside about an hour a day, you can follow it.
By day 30 you will have learned how AI and Claude work, practised prompting and evaluating outputs in the Claude app, set up and maintained a Project, learned the governance rules that apply at work, and sat two full mock exams. CCAO-F format and weights are in registry.yaml and Exam facts.
If you already use Claude regularly, the 15-day plan will save you time.
flowchart LR
W1["Week 1: AI and LLM fundamentals"] --> W2["Week 2: Claude and prompting"]
W2 --> W3["Week 3: output evaluation, products, Projects"]
W3 --> W4["Week 4: governance, workflows, troubleshooting"]
W4 --> F["Days 29-30: mocks and readiness"]What you need¶
- A Claude account. The Free plan is enough for most of the plan; a few exercises mention features (research, chat search) that need a paid plan, and you can read about those instead.
- Optional: Python, if you want to try the labs (they run offline and free in mock mode). See Environment setup. Every lab in this plan is optional; an in-app alternative is always given.
- A notebook or document for your study log: write down every checkpoint question you miss and where the answer lives.
Week 1: AI and LLM fundamentals¶
| Day | Objectives | Reading | Practical exercises | Time | Knowledge checkpoint |
|---|---|---|---|---|---|
| 1 | Understand how the repository and the exam work; choose your plan | Getting started · How to use this repository · Choose your path | Create a Claude account (environment setup: account section); start your study log | 1 h | Explain the four steps of the study loop and the purpose of objective IDs |
| 2 | Explain AI, machine learning and generative AI | What is AI? · Generative AI | In the app, try the classify-versus-generate example on the Generative AI page | 1 h | Both page quizzes fully correct |
| 3 | Explain how an LLM produces text; describe pre-training, fine-tuning and the knowledge cutoff | Large language models | Ask Claude an unfinished sentence and a question about yesterday's local news; note how it handles each. Optional: Lab 01 in mock mode | 1 h | Explain next-token prediction to a friend in three sentences |
| 4 | Define tokens; estimate token counts; explain why output costs more | Tokens and tokenisation | Estimate the tokens in three of your own emails using the rule of thumb; optionally run the page's Python snippet | 1 h | Estimate tokens for a 1,000-word document and explain why long chats cost more |
| 5 | Explain the context window; choose restart, summarise or persist (CCAO-F-3.4) | Context windows | The offsite-planning summarise-and-restart exercise on that page | 1.25 h | Give the right strategy for four scenarios without notes |
| 6 | Explain hallucinations; apply reduction and checking techniques (2.2, 2.3) | Hallucinations · Capabilities and limitations | Ask Claude to summarise a public report with numbers; verify every number against the source | 1.5 h | List five techniques to reduce hallucination and three things to spot-check first |
| 7 | Revision milestone 1 | Re-read any Module 01 page with a missed quiz question | Review the fundamentals flashcards; take a 15-question learning-mode drill on CCAO-F Domain 3 in the mock exam engine (python 18-mock-examinations/run_mock_exam.py --track CCAO-F --domain 3 --mode learn --count 15, or the same in the browser app) |
1 h | 75% or more on the Domain 3 drill; study log updated |
Week 2: Claude fundamentals and prompting¶
| Day | Objectives | Reading | Practical exercises | Time | Knowledge checkpoint |
|---|---|---|---|---|---|
| 8 | Explain how Claude is trained, statelessness and the system, user and assistant roles | How Claude works | The incognito statelessness test on that page | 1 h | Explain why a new chat does not know an old chat's contents, and how Projects and memory change that |
| 9 | Hold a productive conversation; give specific feedback (1.3) | Your first conversation | The full six-step office-move exercise | 1.25 h | Turn three pieces of vague feedback into specific feedback |
| 10 | Write a structured prompt (1.1) | Prompt fundamentals | Rewrite two of your own past requests using the eight building blocks; optional Lab 02 | 1.25 h | List the eight building blocks from memory |
| 11 | Give clear instructions; separate instructions from material | Clear, direct instructions · Structure with XML tags | Wrap a pasted document in tags and ask two questions about it | 1 h | Page quizzes fully correct |
| 12 | Use examples; ask for reasoning when it helps | Examples (few-shot) · Reasoning and thinking | Give Claude two example replies in your team's style and ask for a third | 1 h | Explain when examples help and when they over-constrain |
| 13 | Decompose complex tasks; iterate; adapt to task type (1.2, 1.3, 1.4) | Task decomposition and chaining · Iterating on prompts · Study guide: Domain 1 | Break "plan a team event" into a chain of four prompts; Domain 1 items in practical exercises | 1.5 h | 75% or more on 10 Domain 1 practice questions |
| 14 | Revision milestone 2 | Pages linked from missed questions | Fundamentals flashcards again; 10 more Domain 1 questions | 1 h | 80% or more on Domain 1 overall |
Week 3: output evaluation, products and Projects¶
| Day | Objectives | Reading | Practical exercises | Time | Knowledge checkpoint |
|---|---|---|---|---|---|
| 15 | Choose output formats; adapt outputs for audiences (2.5, 2.6) | Artifacts and output formats · Structured output | The four-formats exercise on the Artifacts page; optional Lab 03 | 1.25 h | Choose a format for five scenarios and justify each |
| 16 | Evaluate accuracy; fact-check; decide when human review is needed (2.1 to 2.4) | Study guide: Domain 2 | Ground an answer in a document with the quote-first prompt from Hallucinations; optional Lab 04 | 1.5 h | 75% or more on 15 Domain 2 questions |
| 17 | Describe the product ecosystem and model families (3.1, 3.2) | Product ecosystem · Model families | Explore your app: find the model menu, the Projects area and the feature menu; note which features your plan has | 1.25 h | Name the four model lines in order of speed and pick a feature for five tasks |
| 18 | Match models to cost, speed and quality (3.3) | Model selection · Study guide: Domain 3 | Run one prompt on two models in the app and compare; optional Lab 06 | 1.25 h | 75% or more on 10 Domain 3 questions |
| 19 | Configure a Project with instructions and knowledge (5.1, 3.4) | Projects and knowledge · What is context engineering? | Build the weekly-update Project from the Projects page; optional Lab 05 | 1.5 h | Explain what goes in instructions versus knowledge, with reasons |
| 20 | Write system-level instructions; manage connectors; maintain configurations (5.2 to 5.4) | System prompts and roles · Connectors and integrations · Memory and persistence · Study guide: Domain 5 | Improve your Project instructions; add a change note; test with three standard questions | 1.5 h | 75% or more on 10 Domain 5 questions |
| 21 | Revision milestone 3 | Pages linked from missed questions | Half-length mock in learning mode covering Domains 1, 2, 3 and 5 | 1.25 h | 75% or more overall |
Week 4: governance, workflows and troubleshooting¶
| Day | Objectives | Reading | Practical exercises | Time | Knowledge checkpoint |
|---|---|---|---|---|---|
| 22 | Describe Anthropic's safety approach (6.1, 6.4) | Anthropic's safety approach | Classify the three example proposals on that page, then three of your own | 1 h | Name the three layers and the two high-risk requirements |
| 23 | Apply data sensitivity and privacy rules (6.2) | Responsible use · Data exfiltration and disclosure | Run the six-question pre-flight checklist on three real tasks | 1.25 h | Explain what to remove before asking Claude to draft a customer reply |
| 24 | Follow organisational policy; understand oversight and injection risks (6.3) | Prompt injection · AI governance · Human oversight · Study guide: Domain 6 | Find and read your organisation's AI policy (or a public example) and note three rules | 1.5 h | 75% or more on 15 Domain 6 questions |
| 25 | Use Claude for requirements, research and design; integrate into workflows (4.1 to 4.4) | Workflows versus agents · Business value and ROI · Study guide: Domain 4 | Map one of your weekly processes into assist, augment and automate steps | 1.25 h | Explain augment versus automate with one example each |
| 26 | Communicate value and limitations (4.5) | Capabilities and limitations (re-read) · Stakeholder communication | Write a half-page note to a manager on what Claude can and cannot do for your team; Domain 4 decision scenarios | 1.25 h | 75% or more on 15 Domain 4 questions |
| 27 | Diagnose and fix poor outputs; optimise workflows (7.1 to 7.3) | Prompts and outputs troubleshooting · Iterating on prompts (re-read) · Study guide: Domain 7 | Fix three weak outputs from earlier in the month with one targeted change each | 1.25 h | 75% or more on 10 Domain 7 questions |
| 28 | Revision milestone 4: full mock exam | None beforehand | Full-length timed mock (python 18-mock-examinations/run_mock_exam.py --track CCAO-F --mode timed, or the browser version) |
2 h | 70% or more in every domain; note your two weakest |
Days 29 and 30: repair and readiness¶
| Day | Objectives | Reading | Practical exercises | Time | Knowledge checkpoint |
|---|---|---|---|---|---|
| 29 | Repair weak areas | Common misconceptions · study guides for your two weakest domains | Retry missed questions (--weak); CCAO-F flashcards |
1.5 h | 80% or more on retried questions |
| 30 | Revision milestone 5: readiness | Revision notes · CCAO-F one-pager · Exam day | Second full mock with a different form; readiness assessment | 2.5 h | Above 80% overall with no domain below 70%, and "ready" on the assessment |
If you are not ready on day 30, that is normal. Add a week: one day per weak domain, then another full mock.
Revision milestones¶
| Day | Milestone | Pass bar |
|---|---|---|
| 7 | Fundamentals flashcards and question set | 75% |
| 14 | Domain 1 consolidation | 80% on Domain 1 |
| 21 | Half-length mock (Domains 1, 2, 3, 5) | 75% |
| 28 | Full timed mock 1 | 70% in every domain |
| 30 | Full timed mock 2 and readiness assessment | 80% overall, 70% per domain |
Practice scores are a study signal only; the questions are unofficial and do not predict an official scaled score.