Prompt Engineering 101: A Complete Beginner's Guide
Prompt engineering is a serious-sounding phrase for a simple skill: writing your request in a way the AI can act on. This guide will teach you the core moves in under fifteen minutes, and give you copy-and-paste patterns that work with any modern assistant.
The one rule of prompts
Give the AI enough context to do the job well. That is the whole rule. Everything below is a specific way to give context.
The five levers you can pull
- Role: who should the AI be?
- Task: what is the actual goal?
- Context: what does it need to know first?
- Format: how should the answer look?
- Constraints: what must it avoid?
A weak prompt vs. a strong one
Weak: "write me an email to my professor".
Strong: "You are a polite computer science student writing to Professor Sharma. Task: request a one-week extension on the OS assignment because of a family emergency. Context: assignment is due Friday, this is my first extension request this semester. Format: 4 short paragraphs, formal but human. Constraints: no exaggeration, no medical details."
The strong version gives you a usable draft on the first attempt.
Chain-of-thought and step-by-step
For anything analytical, ask the model to think step by step. Modern models are trained to do better when you tell them to reason out loud before answering. If you only want the final answer, ask it to reason silently and print only the conclusion.
Few-shot examples
If you can show the model one or two examples of the output you want, do it. Two examples usually outperform a paragraph of instructions.
Iterate, do not restart
The best AI users treat prompts like conversations. Get a draft, then say "shorter", "more formal", "add a joke". That is the whole workflow.
Common mistakes
- Being vague and then blaming the AI.
- Copy-pasting a giant document without saying what to do with it.
- Asking for "perfect" instead of good.
- Not iterating.
Templates you can steal
- Explain X to a smart friend who is not in the field.
- Rewrite the following in the voice of Y.
- List the top 5 things wrong with this, in order of severity.
- What is the strongest counter-argument to this?
- Give me three drafts, from safe to bold.
FAQ
Is prompt engineering going to be obsolete?
The specific tricks change; the core skill of communicating clearly does not.
Do these techniques work on all models?
They work best with modern instruction-tuned models. JARVIS handles all of these patterns well.
Related reading
How to Build Your Own AI Chatbot: A Practical Step-by-Step Guide
A concrete guide for building your first AI chatbot — the architecture, the trade-offs and the exact steps to ship one this weekend.
Web Scraping for AI Research: A Practical Guide
How to responsibly gather data for AI research — techniques, tools, ethics and the mistakes to avoid.