AI Guides › Workbench
By Nigel Guy · 6 min read
Most people collect AI courses the way they collect browser tabs: twenty started, none finished, and a CV line that reads "completed various AI courses". It feels like progress because each certificate is a small win. A recruiter sees a list with no evidence that you can do anything with it.
The rule: take few courses, finish them, and attach one piece of real work to each credential you list.
All of the below are free to start at time of writing. Check each page before you commit, because catalogues change often. Where a source could not be checked directly, the Catch column says so.
| Course | What it does | Cost | Best for | Catch |
|---|---|---|---|---|
| Claude 101 (Anthropic) | Everyday work with Claude: core features and workflows | Free | Absolute beginners on Claude | Short; shows familiarity, not skill |
| AI Fluency: Framework & Foundations (Anthropic) | A structured way of working with AI: delegating, describing, discerning, diligence (Anthropic's "4D" framework) | Free | Anyone who uses AI at work | Concepts, not tool clicks |
| AI Capabilities and Limitations (Anthropic) | How these models work and where they fail | Free | Spotting errors and over-trust | Introductory |
| Claude Code 101, Claude Code in Action, Introduction to agent skills, Introduction to subagents (Anthropic) | Coding-agent workflows | Free | Developers and technical writers | Assumes you code or will learn |
| Building with the Claude API; Introduction to Model Context Protocol (Anthropic) | Building on the API; building MCP servers and clients in Python | Free | Builders | Needs Python and an API account, which has its own usage costs |
| OpenAI Academy | Tracks for work, building, leading adoption and teaching | Pricing not stated on the page I could read | Knowledge workers, educators | See "What I could not confirm" below |
| Elements of AI (University of Helsinki and MinnaLearn) | What AI is and is not; light on tools | Free | Sceptics wanting grounding | A separate "Get AI Ready" strand offers a certificate; the main courses do not mention one |
| Hugging Face Learn | LLM Course, Agents Course and others | Free | Technical readers | Hands-on and code-heavy |
| DeepLearning.AI | Short courses, including generative AI for non-specialists | Free or paid status not shown on the page I read; check before you start | Structured short learning | Membership tiers exist |
Anthropic's catalogue also includes role-specific AI Fluency courses (for educators, students, nonprofits, small businesses, builders, creative work). Pick the one that matches your job, not all of them.
This is the mechanism. One row per course, filled in before you add anything to a CV.
| Course | Finished on | One thing I made or changed because of it | Where the proof lives |
|---|---|---|---|
| Example row | [DATE] | [A rewritten weekly report process, a script, a checklist] | [Link, folder or portfolio page] |
A course earns a CV line only when the third and fourth columns are filled in. Otherwise it stays in your notes.
Fill in your role, hours per week and goal. Paste it into the assistant you already use.
You are a careful learning adviser for working professionals in the UK.
Context: I work as [JOB_ROLE] in [INDUSTRY]. I can spend [HOURS_PER_WEEK] hours a week for [NUMBER_OF_WEEKS] weeks. My goal is [GOAL, e.g. a promotion, a job change, using AI in my current role]. My coding level is [NONE / BASIC / COMFORTABLE].
Task: from the free courses I list below, choose at most three and order them. Explain in one sentence why each is included and what you left out and why.
Courses available to me: [PASTE_COURSE_NAMES_AND_LINKS_YOU_HAVE_CHECKED].
Output: a table with columns Course, Week to finish, A real task from my job to apply it to, Evidence I will keep. Then a two-line CV wording template for each course with blanks for my results.
Rules: use only the courses I listed; do not invent course content, durations or certificates; if you are unsure what a course covers, say so and tell me what to check on its page. If any input above is missing, ask me before answering.
Before you answer, check that the plan fits my weekly hours and that every course has a task and a piece of evidence.
| If you are... | Start with | Then |
|---|---|---|
| New to AI at work | Claude 101 | AI Fluency: Framework & Foundations |
| A manager or team lead | OpenAI Academy's adoption track, or Building effective human-agent teams (Anthropic, listed as beta) | AI Capabilities and Limitations |
| A teacher | AI Fluency for educators, or OpenAI's teacher course | Your own lesson plans as evidence |
| A developer | Claude Code 101 | Building with the Claude API |
| Sceptical and want grounding first | Elements of AI | One tool course |