A Beginner's Course for the AI Age
Last updated: October 2026
In today's world, artificial intelligence is everywhere — from virtual assistants to creative tools, from research helpers to coding companions. Prompt engineering is the new literacy of the AI age. It's the skill that lets you communicate with AI tools like Google Gemini clearly enough to get genuinely useful results.
Whether you're writing essays, studying for exams, brainstorming business ideas, or solving complex problems, knowing how to prompt well is like having a superpower. This course will take you from complete beginner to confident user in four weeks — using only free tools.
A prompt is any instruction, question, or piece of text you give to an AI. Prompt engineering is the practice of crafting those instructions intentionally to get the best possible output.
Think of it like giving directions. If you tell someone "go to the store," you might get a confused look. But if you say "Turn left out the driveway, drive three blocks, and stop at the Walmart on the corner," they'll get there every time. Same principle with AI.
The core truth: The quality of your output is almost entirely determined by the quality of your input.
Before diving into assignments, understand these seven principles. They're the foundation of everything.
Vague prompts produce vague answers. The more specific you are, the better the output.
AI doesn't know your situation unless you tell it. A few sentences of context dramatically improve results.
Assigning a role to the AI shapes how it responds. Use "Act as a..." to unlock specialized expertise.
Tell the AI exactly how you want the information presented. Tables, bullet points, numbered lists, outlines — the AI can produce any format.
No one writes the perfect prompt on the first try. Treat it like a conversation. Refine based on what you get back.
Few-shot prompting is dramatically more powerful when you need a specific style or format.
For complex problems, tell the AI to reason through them rather than jump to an answer.
This simple instruction often doubles or triples the quality of the answer on hard questions.
Goal: Get comfortable with Gemini's interface, understand what prompts are, and practice the basics.
Goal: Get familiar with Gemini and observe how different prompt styles produce different results.
Open Gemini and try these three prompts back to back:
Observe: How dramatically different are the outputs? Which is most useful and why?
Goal: Experience firsthand how specificity changes output quality.
Pick any topic you're interested in (your major, a hobby, a current event). Ask Gemini about it two ways:
Write down: What changed? What would you tell a friend about "being specific" based on what you saw?
Goal: Practice assigning roles to unlock different expertise.
Try these role-based prompts in Gemini:
Observe: How does the role affect the tone, depth, and type of advice?
Goal: Learn to control the shape and structure of AI output.
Ask Gemini the same question in three different formats:
Which format would you actually use? Why?
Goal: Master iteration, few-shot prompting, and chain-of-thought. Start using AI for real college tasks.
Goal: Practice refining prompts through multiple rounds.
Start with this prompt: "Help me write an introduction paragraph for an essay." Then iterate at least 4 times by adding more context, changing the tone, specifying the topic, adjusting length, etc.
Document each version of your prompt and the output. By round 4, you should have something you'd actually consider using.
Reflect: What did you add/change each round and why?
Goal: Teach the AI your style with examples.
Imagine you write in a specific casual style. Give Gemini 2–3 examples of that style, then ask it to write something new in that same voice.
Try it with: Instagram captions, text messages to a friend, or your own essay writing style.
Prompt template: "Here are three examples of how I write captions: [example 1], [example 2], [example 3]. Now write a caption for a photo of [describe photo] in the same style."
Observe: How close does the AI get to your actual voice?
Goal: See how "think step by step" improves complex reasoning.
Give Gemini a logic puzzle, a math word problem, or a decision you're actually facing. Ask it twice:
Compare the two responses. Is the step-by-step version more accurate or trustworthy?
Goal: Apply prompt engineering to a real academic task.
Pick a subject you're currently studying or want to learn. Ask Gemini: "Act as an expert professor in [subject]. I'm a college freshman. Create a one-page study guide for the topic of [specific topic]. Include: key vocabulary with definitions, 3 core concepts explained simply, 5 review questions, and one common misconception students have."
Iterate: Adjust the level of difficulty, add or remove sections, ask it to make it more visual or concise.
Goal: Push into more sophisticated techniques — personas, prompt chaining, and using AI as a thinking partner rather than just an answer machine.
Goal: Break complex tasks into a chain of connected prompts.
Research a topic in 4 chained steps in the same Gemini conversation:
Notice how each prompt builds on the last. This is prompt chaining — one of the most powerful techniques.
Goal: Use AI as an intellectual sparring partner.
Pick a topic you have an opinion on. Ask Gemini: "I believe [your position]. Act as a smart, respectful devil's advocate and give me the strongest possible arguments against my position. Then tell me which of my counterarguments you think holds up best."
This is one of the most powerful study/critical-thinking uses of AI — not to confirm your beliefs, but to stress-test them.
Goal: Learn AI's biggest weakness — making things up.
Ask Gemini about a very specific factual topic (a historical date, a scientific study, a legal detail). Then fact-check its answer using Google or Wikipedia.
Did it get everything right? Did it make up any details? Write a short reflection: When should you trust AI outputs and when should you always verify?
Goal: Build a persistent character for a creative or professional task.
Ask Gemini to act as a specific persona for an extended conversation. Examples:
Have a full back-and-forth conversation. Observe how well the AI maintains the persona and how to correct it when it breaks character.
Goal: Apply everything you've learned to real-world college tasks. By the end of this week, you should be using Gemini as a regular part of your workflow.
Goal: Use AI to help (not replace) your writing process.
Take an essay topic from one of your classes (or make one up). Use Gemini to:
Important: The AI is a collaborator, not the author. Your thinking drives it. Does the final paragraph sound like you?
Goal: Write professional emails faster.
Think of a real email you need to send (to a professor, a club, a future employer, a landlord). Ask Gemini: "I need to send an email to [recipient] about [situation]. The tone should be [professional/friendly/formal]. Key points I want to include: [list]. Draft this email in under 150 words."
Edit the draft to sound more like you. What did you change?
Goal: Turn any reading into instant study materials.
Take a page or two from a textbook, article, or lecture notes. Paste it into Gemini and ask: "Convert this content into 10 flashcard-style Q&A pairs. Format each as: Q: [question] / A: [answer]. Focus on the most testable concepts."
Try this with actual upcoming exam material. How long did it take vs. making them manually?
Goal: See how non-coders can use AI to automate small tasks.
Ask Gemini to write a simple script or formula for something practical:
You don't need to understand all the code. The goal is to see that AI makes technical tools accessible even to non-technical people.
Part 1 of the AI Fundamentals Series. Uses Google Gemini (free) for all assignments.
Next: Scaffolding 101 →