← BackGrades 10-12 | Chapter 1: AI Tools

Prompt Engineering 🎯

Getting Better AI Results

Session 2: The Art of AI Prompts

What Is Prompt Engineering? 🤔

Prompt engineering is the skill of writing effective instructions for AI to get the best possible outputs. It's become a real job skill in tech companies.

Why It Matters:

  • Bad prompts = bad AI responses
  • Good prompts = excellent AI responses
  • Same AI, vastly different results based on how you ask
  • Companies hire "prompt engineers" to make AI useful
  • This is a skill you can learn and master!

Bad Prompts vs Good Prompts 📝

Bad Prompt:

"Write an essay about climate change"

Good Prompt:

"Write a 3-paragraph persuasive essay about climate change from the perspective of a high school student applying to college. Focus on: (1) the problem, (2) human responsibility, (3) solutions. Use scientific evidence. Aim for 500 words. Use an academic tone suitable for college applications."

The difference? Specificity. The AI knows exactly what you want!

Prompt Engineering Techniques 🔧

Technique 1: Be Specific

Instead of "write about AI", say "Explain how neural networks work in 3 paragraphs for someone with no tech background"

Technique 2: Give Context

Tell AI who you are, what you need it for: "I'm a high school student writing a research paper on AI ethics"

Technique 3: Request Specific Format

Instead of "write something", specify: "Create a 5-bullet-point summary" or "Format as a professional email"

More Prompt Techniques 🔧

Technique 4: Use Examples

Show the AI an example of what you want: "Write a professional email like this example: [insert good example]"

Technique 5: Specify Tone & Style

Tell AI how to sound: "Use a formal, professional tone" or "Write this like you're talking to a friend but keep it professional"

Technique 6: Give Constraints

Set limits: "Use simple language, no jargon" or "Keep it under 100 words" or "No references to politics"

Technique 7: Iterate & Refine 🔄

Your first prompt won't be perfect. Ask follow-up questions to improve!

Example Iteration:

  1. First prompt: "Write about climate change"
  2. AI response: Generic, not what you wanted
  3. Follow-up: "That's good but make it more specific to ocean acidification"
  4. AI refines it
  5. Follow-up: "Can you add data from 2023?"
  6. Keep refining until it's perfect!

Advanced: Role-Playing with AI 🎭

Tell the AI to play a role. This often improves responses.

Examples:

  • "You are a college admissions counselor. Critique my essay and suggest improvements"
  • "You are a professional recruiter. Critique my resume for a tech job"
  • "You are a science teacher. Explain quantum physics to a 10th grader"

Why it works: The AI adjusts its knowledge, tone, and style to fit the role!

Chain of Thought Prompting 🧠

Ask the AI to "think step by step". This improves accuracy and reasoning.

Example:

Bad: "Is climate change real?"

Better: "Think step by step about the scientific evidence for climate change. What are 5 key pieces of evidence?"

The AI will provide more thoughtful, detailed reasoning!

Hands-On Activity: Prompt Competition 🏆

Challenge: Write the best prompts to get the best AI outputs.

Tasks:

  1. Professional Email: Create a prompt to get the best professional email asking for a recommendation letter
  2. Concept Explanation: Write a prompt to explain a difficult concept (pick any subject)
  3. Study Guide: Create a prompt for a study guide for an exam

Judging: Which prompt produced the best, most useful response?

Common Prompt Mistakes ❌

Avoid These:

  • Too vague: "Write something" - AI doesn't know what you want
  • Contradictory instructions: Asking for both very short AND comprehensive
  • Not giving context: AI doesn't know your audience or purpose
  • Asking AI to do something it can't: "Tell me what will happen in 2050" (speculation, not fact)
  • Not iterating: Accepting first response without refinement

Real-World Prompt Engineering Jobs 💼

Companies Hiring Prompt Engineers:

  • OpenAI, Anthropic, Google: AI companies
  • Amazon, Microsoft, Meta: Tech companies with AI products
  • Startups: Using AI to build tools

Salary Range:

Entry-level: $80k-$120k/year (strong competition)

Senior: $150k-$200k+/year

Required: Strong communication skills + understanding of AI + critical thinking

What We Learned 🎓

  • Prompt engineering is the skill of writing effective AI instructions
  • Specificity matters: Detail, context, constraints = better results
  • Techniques: Be specific, give context, specify format, use examples, iterate
  • Role-playing: Make AI play a role for better responses
  • Chain of thought: Ask AI to explain step-by-step
  • This is a real job skill with strong demand and good pay

Prompt Master! 🎉

You can now get excellent results from AI tools

Next Session: AI in Different Career Fields