Lesson library

Practical AI lessons for real life

Free lessons include the full walkthrough. Premium lessons show a safe preview here — sign in for practice, quizzes, and spaced reviews inside Vibed.

🌱 Free course · 8 lessons

Your First Week with AI

Never used an AI assistant? Seven short days from opening the tool for the first time to a ten-minute daily habit — built for people who feel the office moving without them.

View course

🧭 Free course · 12 lessons

AI at Work Foundations

What AI tools can and cannot do, when to trust them, and how to get useful answers on the first try.

View course

1. What AI actually is (and is not)

An LLM predicts plausible text from patterns; it is not a database of facts

2. Your first real prompt: context, task, format

A useful prompt states the context, the task, and the output format

3. When to trust the answer

Verify anything the model recalled on its own; trust transformations of material you supplied

4. Give it a role

Assigning the model a role focuses its expertise and tone on your situation

5. Show, don't tell: examples beat descriptions

One or two examples of the output you want beat paragraphs of description

6. The critique loop: make it improve its own work

Asking the model to critique and revise its own output reliably raises quality

7. Ask better follow-up questions

The first answer is a starting point; targeted follow-ups that name what to change beat starting over

8. Make AI ask you questions first

When your request is fuzzy, tell the AI to ask you questions before it answers

9. Use your own facts first

Supply the facts yourself and ask the model to do the language work, not the recall

10. Check before you send

Run a quick verification pass over numbers, claims, and tone before forwarding AI-assisted work

11. Private, sensitive, or safe to paste?

Classify information as public, internal, or confidential before putting it into AI

12. Your personal AI task map

Choose where to use AI by mapping your tasks on value/frequency against risk

💬 Premium course · 10 lessons

Claude Mastery

Projects, file analysis, drafting workflows, and artifacts — turn Claude into your sharpest colleague.

View course

🧩 Premium course · 12 lessons

Prompting Patterns

Six repeatable patterns that work in any AI tool — decompose big jobs, make it interview you, compare options, and steer without starting over.

View course

1. One job per prompt

Big bundled requests produce mush; a pipeline of small single-job prompts produces quality

2. Make it interview you

Telling the model to ask you questions first beats guessing what context it needs

3. Options first, then commit

Asking for three distinct versions with trade-offs beats accepting the first answer

4. Constraints are instructions too

Stating hard limits — length, banned content, must-keep facts — prevents the failures you cannot un-see

5. Steer with deltas, restart when stuck

Targeted change requests beat re-explaining, and a fresh chat beats a polluted one

6. Build your prompt library

A winning prompt with fill-in-the-blank slots is a tool your whole team can reuse

7. Success criteria before prompt wording

Define what good looks like before you ask the model to produce anything

8. Plan, do, check

Ask for a plan, execute one step, then verify against criteria before continuing

9. Extract, transform, apply

Turn messy input into structured output, then use that structure in the next prompt

10. Red-team your answer

Ask a second pass to find risks, assumptions, weak claims, and missing perspectives

11. Structured outputs: tables, JSON, checklists

Format is part of the task — specify columns, keys, or checklist items exactly

12. Prompt repair clinic

Diagnose weak prompts with a checklist: job, context, criteria, format, scope

🏪 Free course · 10 lessons

Small Business AI Foundations

Ten practical lessons for owners who need more customers, faster replies, better admin, clearer money, and safer AI habits.

View course

📈 Premium course · 16 lessons

Small Business Growth & Operations System

A Pro course for turning AI into repeatable workflows: marketing calendars, offers, follow-ups, quotes, cashflow reviews, SOPs, and owner playbooks.

View course

1. Build your business command centre

Keep business facts, offers, prices, policies, and tone in one reusable brief

2. The customer journey map

Map how strangers become repeat customers

3. Create a 30-day marketing calendar

Marketing becomes easier when themes, offers, and channels are planned

4. Write offers people understand

Good offers say who it is for, what they get, why now, and what to do

5. Local business content engine

Turn real shop moments into posts: before-after, FAQ, customer story, behind the scenes

6. WhatsApp sales follow-up

Most sales need reminders, not pressure

7. Customer complaint recovery

A calm complaint response can save reputation and reveal operational problems

8. Quote and proposal builder

A quote should make scope, price, timing, exclusions, and next step clear

9. Supplier comparison without spreadsheets

AI can compare offers if you provide the facts and criteria

10. Pricing sanity check

AI can help think through costs, margin, positioning, and customer value

11. Weekly cashflow conversation

Owners need a weekly habit to understand money in, money out, and risks

12. Stock and inventory notes

AI can help spot fast movers, slow movers, and reorder risks from simple notes

13. SOPs from how we do it

Turn owner knowledge into repeatable staff instructions

14. Staff training cards

Small teams need clear, short, repeatable training

15. Monthly owner review

Review customers, cash, operations, risks, and next experiments

16. Your small-business AI playbook

Save the best workflows as reusable templates

💼 Free course · 10 lessons

AI Career Foundations

Ten practical lessons for job seekers and workers who need proof-based CVs, stronger applications, better interviews, and safer job-search habits.

View course

🚀 Premium course · 16 lessons

Career & Income Accelerator

A Pro course for building a serious career system: target roles, evidence inventory, tailored CVs, portfolio samples, interview practice, and monthly reviews.

View course

1. Choose a realistic target role

Better applications start with a clear target

2. Your career evidence inventory

Collect proof from work, school, volunteering, family business, projects, and life

3. Translate experience into employer language

Employers buy outcomes, not duties

4. Build a master CV

One strong master CV feeds tailored versions

5. Tailor a CV to a job ad

Match relevant proof to the role without keyword stuffing or lying

6. Write human cover notes

Short, specific messages beat generic AI letters

7. Build a work-sample portfolio

Proof can be created through sample work, not only past jobs

8. Networking without begging

Good outreach asks for advice, information, or a small next step

9. Interview practice with AI

AI can simulate interviews, but you must answer from real stories

10. Difficult interview questions

Prepare for gaps, failures, salary, conflict, and weaknesses

11. Salary and offer conversations

Prepare calmly using value, market context, and priorities

12. Your 30-day skill sprint

Learn one valuable skill through small daily proof-building tasks

13. Use AI once you get the job

AI can help you onboard, summarize, plan, and communicate

14. Freelance or side-income basics

A small service offer can create experience and income

15. Your career dashboard

Track applications, conversations, interviews, skills, and follow-ups

16. Monthly career review

Review evidence, gaps, applications, and next moves

⚖️ Free course · 12 lessons

AI Legal Research Foundations

A free safety-first course for non-lawyers: legal information versus advice, SOURCE method, jurisdiction, source verification, lawyer briefs, and red flags.

View course

1. Legal information is not legal advice

AI can explain legal information, but it cannot replace a lawyer's judgment

2. The SOURCE method

Research starts with facts, jurisdiction, official sources, validation, and escalation

3. What law applies where?

Jurisdiction matters: South African law is not US, UK, or Australian law

4. Find the legal issue, not just the emotion

“This is unfair” must become a legal research question

5. Use AI to build a legal research plan

AI is useful for planning searches and keywords, not as the final authority

6. Sources of law in plain English

Constitution, Acts, regulations, by-laws, cases, court rules, contracts, and policies are different sources

7. Cases and precedents in plain English

A case matters because of the rule it decides, not because it sounds similar

8. Never trust a case until you verify it

Check the case exists, citation, court, date, paragraph, and proposition

9. Read legislation without panic

Definitions, scope, exceptions, commencement, amendments, and regulations matter

10. Ask AI to explain a case you provide

Paste the real judgment and ask for issue, rule, facts, outcome, and caveats

11. Build a legal question for a lawyer

The best use of AI may be preparing a clear attorney brief

12. When to stop researching and get help

Deadlines, court papers, threats, arrests, eviction, custody, and money claims need human help

🇿🇦 Premium course · 16 lessons

South African Legal Research with AI

A Pro course for careful South Africa-grounded research using primary sources, court hierarchy, SAFLII, LawLibrary, legislation, cases, and authority tables.

View course

1. The South African legal source map

Know where to look: Constitution, Acts, regulations, gazettes, cases, rules, regulators

2. Court hierarchy and precedent

Authority depends on court level, issue, and whether a decision binds or persuades

3. Use SAFLII safely

Search, open, read, and cite South African cases from primary sources

4. Use LawLibrary safely

Search judgments, legislation, gazettes, and recent legal materials

5. Legislation currentness

Acts change; check amendments, commencement, regulations, and version dates

6. Read a judgment like a researcher

Identify facts, issue, rule, reasoning, order, and paragraphs worth citing

7. Ratio, obiter, and outcome

Not every sentence in a judgment is the legal rule

8. Binding vs persuasive authority

Higher courts bind; similar, lower, or foreign decisions may persuade but need caution

9. The authority table

Every legal claim must connect to a source, paragraph, and verification status

10. Validate an AI legal answer

Break AI output into claims and verify each claim independently

11. Statutes plus cases

Legislation gives the rule; cases show how courts interpret and apply it

12. Foreign law traps

UK, US, Australian, and EU law may be useful background, but not South African authority

13. Legal research memo

A good memo states question, sources, answer, uncertainty, and next steps

14. Preparing for attorney consultation

Give the lawyer facts, documents, questions, risks, and what you already checked

15. Research for disputes before court

Organize facts, documents, evidence, timelines, and possible legal routes

16. Final project: verified research pack

Demonstrate safe legal research without giving yourself legal advice

📄 Premium course · 12 lessons

Legal Research for Small Businesses and Teams

A Pro/Team course for contracts, supplier disputes, debt, labour, POPIA, consumer complaints, leases, risk registers, and attorney-ready briefs.

View course

🛡️ Free course · 10 lessons

AI Fraud Awareness Foundations

Free prevention: spot AI-powered scams before you click, pay, share, approve, or panic.

View course

⏱️ Free course · 10 lessons

If Money Is Gone — The First 24 Hours

Free emergency course: limit damage, preserve evidence, and start your case file in the first day.

View course

🔍 Premium course · 12 lessons

Bank Fraud Investigation for Non-Technical People

Pro course: test the bank’s rejection with timelines, records, and the PROVE framework.

View course

📋 Premium course · 10 lessons

Data Requests, POPIA, PAIA, and Bank Records

Pro course: request the right bank records lawfully and clearly.

View course

📨 Premium course · 12 lessons

Build the Bank Complaint and NFO Pack

Pro course: affidavits, complaint letters, rejection responses, and NFO escalation.

View course

🏢 Premium course · 12 lessons

Small Business and Workplace Bank Fraud Defence

Team course: invoice fraud, supplier changes, dual approvals, and business NFO packs.

View course

🛠️ Free course · 8 lessons

Vibe Coding Foundations

Build your first app with AI — from a small idea and brief to prototype, review, and production readiness.

View course

🚀 Premium course · 4 lessons

Vibe Coding Builder

Your first production app path — mindset, prototype, workbench setup, and app shell.

View course

⚖️ Premium course · 8 lessons

AI Safety & Professional Judgment

Classify risk, handle data safely, spot false confidence, and build team guardrails — not one lesson, a habit.

View course

📄 Premium course · 8 lessons

Documents & Research

Source-first habits: evidence chains, decision briefs, contradictions, and number verification.

View course

✍️ Premium course · 8 lessons

AI for Writing & Communication

Write for the reader: executive summaries, difficult emails, memos, proposals, and voice guides.

View course

📅 Premium course · 8 lessons

Meetings, Admin & Operations

Save hours weekly: meeting actions, SOPs, checklists, support macros, reports, and handovers.

View course

📊 Premium course · 8 lessons

AI for Excel, Sheets, and Numbers

Spreadsheet judgment: clean text, ask business questions, summarize tables, formulas, pivots, anomalies, charts, verify totals.

View course

🚀 Premium course · 12 lessons

ChatGPT & Codex Mastery

Get real work done with ChatGPT — projects, deep research, data analysis, and agents — then delegate your first small build to Codex without becoming a developer.

View course

1. The ChatGPT map: modes matter more than models

Fast answers, thinking, deep research, and agent mode are different tools — picking the mode is the skill

2. Memory and custom instructions that work for you

Teach ChatGPT your role, formats, and standards once — and control what it remembers

3. Projects: one workspace per recurring job

A project bundles instructions, files, and memory so recurring work starts pre-briefed

4. Deep research: a cited brief while you make coffee

Deep research reads dozens of sources and returns a cited report — your job is scoping and verification

5. Canvas: draft documents side by side, not in the chat stream

Canvas separates the document from the conversation so you can edit surgically instead of regenerating

6. Data analysis: tables in, checked answers out

ChatGPT runs real Python on your uploaded data — powerful, and only trustworthy with spot checks

7. Agent mode with a safety harness

Agents click, browse, and complete multi-step tasks — supervision and boundaries are what make that safe

8. Scheduled tasks: AI on a weekly rhythm

ChatGPT can run recurring tasks unattended — worth it only for briefings, never for unchecked decisions

9. Your data: consumer ChatGPT vs work plans

Where you type decides where your words can go — plan type and settings are a professional duty

10. Codex: what it is and why non-developers should care

Codex is an agent that writes and runs code from a plain-language brief — the brief is your job

11. Your first Codex delegation: brief a tiny tool

A buildable brief has purpose, inputs, outputs, rules, and a test you can run without reading code

12. Codex automations, skills, and knowing when to hand over

Standing agents multiply both value and blast radius — start read-only, review rhythms, know the handover line

📎 Premium course · 10 lessons

Microsoft Copilot Mastery

Master Microsoft 365 Copilot where you already work — Word, Excel, Outlook, Teams, and PowerPoint — grounded in your own files and checked before anything ships.

View course

1. The Copilot map: Chat, in-app Copilot, and agents

Copilot is three layers — Chat, in-app helpers, and agents — and web versus work grounding decides what it knows

2. Ground it in your work: files, people, and meetings

Referencing specific files, people, and meetings turns Copilot from a generic writer into YOUR analyst

3. Word: a first draft from your sources, not from vibes

Draft from referenced documents, edit by selection, and keep names, numbers, and commitments human-checked

4. Excel: ask your table a business question

Copilot analyses, explains formulas, and follows workbook rules — and you still spot-check the totals

5. Outlook: summarise threads, draft replies, keep your name on it

Thread summaries, prioritisation, and selection rewrites cut email hours — sending stays a human act

6. Teams meetings: recap, actions, and honest minutes

Meeting recap turns transcripts into decisions and owners — consent and correction stay human work

7. PowerPoint: a deck from a document, not from vibes

Decks built from a source document inherit real content — you own the narrative and every number on screen

8. Agent mode and productivity agents: supervise like a manager

Copilot agents take multi-step actions in your documents — checkpoints and review turn power into safety

9. Copilot Pages and your prompt library: make good prompts reusable

Pages turn chat answers into living team documents; a prompt library turns your best prompts into assets

10. Governance: what Copilot can see is what it can surface

Copilot inherits your permissions — oversharing, labels, and knowing when to escalate are part of using it well

🧭 Premium course · 10 lessons

AI for Managers & Team Leads

Your team is already using AI — with or without you. Set the ground rules, review AI-assisted work fairly, delegate across people and machines, and grow the skills without a training budget.

View course

🎧 Premium course · 10 lessons

AI for Customer Support & Sales

Front-line AI that customers never regret: replies in your company voice, calm de-escalation drafts, clean handovers, sharper call prep, and follow-ups that remember the conversation.

View course

1. Where AI helps the front line — and where it burns trust

AI compresses the writing and reading; the customer relationship stays handmade

2. Replies that sound like your company, not like AI

Two example replies teach the voice better than any tone adjective

3. The angry-customer reply

AI keeps the draft calm at scale — the promise-check and the sincerity stay yours

4. The clean handover: case summaries in ninety seconds

A structured case summary transfers a customer without making them repeat themselves

5. Turn solved tickets into team knowledge

Ten minutes after a hard-won solution, AI turns it into an article the whole team can reuse

6. Ten-minute research before the sales call

A structured research brief beats an hour of tab-hopping — with claims verified before you repeat them

7. Follow-ups that remember the conversation

Call notes in, personal follow-up out — with every commitment checked against what you actually said

8. Proposals in an afternoon, numbers checked twice

Assemble proposals from your real sources at AI speed — prices and promises verified by hand

9. Customer data is not AI fuel

The trust customers place in the front line creates duties no deadline suspends — strip, minimise, and use approved tools

10. Your front-line playbook

Assemble the course into a daily loop with saved prompts — and measure the minutes it buys

🔷 Premium course · 10 lessons

Gemini & Google Workspace Mastery

Master Gemini where your work already lives — Gmail, Docs, Sheets, Slides, and Meet — plus NotebookLM for grounded research and Gems for reusable assistants, checked before anything ships.

View course

1. The Gemini map: chat, side panels, and your data

Gemini lives in three places, and what it knows depends on where you ask and what you reference

2. Gmail: digest threads, draft from bullets, read before send

Gemini summarises the thread and drafts the reply — the send stays a human act

3. Docs: first drafts from your own files

Help-me-write grounded in Drive files — with style matching — beats generation from nothing

4. Sheets: formulas explained, tables questioned, totals verified

Gemini writes and explains formulas and answers table questions — one spot-check makes it trustworthy

5. Slides: the deck comes from the document

Generate slides from a Doc that says something — you own the narrative and every number on screen

6. Meet: automatic notes, human-checked minutes

Take-notes-for-me turns meetings into records — consent and a correction pass keep the record honest

7. NotebookLM: answers only from sources you chose

Upload the documents; get cited answers from them and nothing else — verification is one click

8. Gems: brief once, reuse forever

A Gem is a saved assistant with your standing instructions — one per recurring job

9. Data boundaries: Gemini sees what you can see

Grounded AI inherits your permissions — surprises are oversharing bugs, and confidential stays labelled

10. Your Google-stack daily loop

Wire the nine skills into the day you already have — and know when to reach outside the stack

Reading is recognition. Practice is the skill.

Members use each idea against a real AI assistant, get feedback on the prompt, and see the lesson again before it disappears from memory.

Start practicing free