Back

Intermediate

Master prompt engineering, work with documents and images, and build reliable AI workflows.

HK$6,000

Detailed 5-day curriculum

  1. Day 1: Prompt engineering foundations

    The R-C-C-F framework

    • Role — 'You are a patient tutor'
    • Context — background information
    • Constraints — length, tone, format
    • Format — bullet list, table, email

    Everyday comparison

    • Instead of: Write something about my business.
    • Try: You are a marketing copywriter. Context: I sell handmade candles in HK. Constraint: 100 words, warm tone. Format: Instagram caption with 3 hashtags.

    Hands-on practice

    • Rewrite 3 weak prompts using R-C-C-F. Share one with the class.

    Knowledge check

    • Which element tells AI HOW to structure the output?

    Safety habit

    • Even well-crafted prompts should not include others' private data.
  2. Day 2: Iteration & critique

    The iteration loop

    • Generate → Read critically → Refine prompt → Compare versions
    • Ask: 'What assumptions did you make?'
    • Ask: 'Give me 3 alternatives'

    Hands-on practice

    • 3-round challenge: Start with a vague prompt, improve twice, document what changed each round.

    Analogy or example

    • Tap for refinement phrases
    • 'Shorter', 'More formal', 'Add HK-specific examples', 'Remove jargon', 'Show your reasoning step by step'

    Knowledge check

    • First output isn't great. Best next step?

    Safety habit

    • Critique AI outputs before sharing them publicly or with clients.
  3. Day 3: Working with documents & images

    Document workflows

    • Upload PDFs (check privacy settings first)
    • Ask for summaries, key points, action items
    • Extract data into tables

    Hands-on practice

    • Summarise a sample school newsletter: 3 key dates, 1 action for parents, tone check.

    Image basics

    • Generate images from text descriptions
    • Edit: 'remove background', 'make brighter'
    • Always disclose AI-generated images when sharing commercially

    Knowledge check

    • Before uploading a document to AI, you should…

    Safety habit

    • Redact names and account numbers from documents before upload.
  4. Day 4: Productivity stack

    Daily productivity uses

    • Email drafts & replies
    • Meeting notes → action items
    • Spreadsheet formulas explained
    • Calendar planning prompts

    Hands-on practice

    • Automate one real admin task you do weekly. Write the prompt and test it live.

    Knowledge check

    • Best practice for email drafts?

    Safety habit

    • Never auto-send emails containing financial or legal commitments without human review.
  5. Day 5: Fact-checking & verification

    Trust but verify

    • Cross-check facts with official sources
    • Ask AI for sources — then click them
    • Use multiple tools for important research

    Hands-on practice

    • Verify 5 AI-generated 'facts'. Mark each TRUE, FALSE, or UNVERIFIABLE with your source.

    Take-home

    • Build your prompt library — 10 saved prompts for home and side-hustle tasks.

    Knowledge check

    • AI cites a study from 2019. You should…

    Safety habit

    • Health, legal, and financial advice from AI always needs professional verification.