Prompt Engineering

Anatomy of a good prompt

Most bad AI answers come from vague prompts. Learn the six parts of a strong prompt and turn a one-line request into one that gets a useful answer the first time.

Be specific: context, constraints and audience

Turn a vague request into a precise one by adding context, measurable constraints and a clear audience, one layer at a time, and see how each layer changes the answer.

Zero-shot and few-shot: teaching by example

Learn when a plain instruction (zero-shot) is enough and when to show a few examples (few-shot). You'll build a review classifier and a style-matching prompt, and learn how to choose examples that teach the right thing.

Getting structured output: tables, JSON and checklists

Learn to ask for answers in a fixed shape (tables, checklists and JSON) so you can scan them quickly, paste them into other tools or read them in code. You'll also see how to check and repair JSON that comes back.

Step-by-step reasoning prompts

Learn why asking an AI to work through a problem in steps gives better answers on maths, logic and decisions, and how to check its working, compare options and avoid being fooled by confident but wrong reasoning.

Prompt chaining: breaking big tasks into steps

Learn to split a big task into a chain of small prompts, where each answer feeds the next. You'll plan a Brew & Bean weekend promotion by hand, then wire the same chain into a short Python script.

Iterate: critique, refine and retry

Your first answer is a draft. Learn a practical loop for improving it: give specific feedback, ask the model to critique itself, change one thing at a time and know when to start a fresh chat instead of pushing on.

System prompts, custom instructions and personas

Write the standing instructions an AI follows in every reply: a system prompt for a small shop assistant, custom instructions for your own chats, and personas for practice. Then test it and run it from Python or a local model.

Working with long documents and files

Learn how to get accurate answers from long PDFs, reports and notes: how context windows work, how to ask questions that stay grounded in the text, and how to split a document that is too big.

Prompting image and media generators

Learn a simple formula for prompting AI image generators, how to iterate on a result, and what they cannot do reliably. Includes ready prompts for posters, logos, thumbnails and diagrams, and the rules for using AI images responsibly.

Prompts for studying and exam preparation

Use AI as a tutor instead of an answer machine: explain hard topics at your level, build a revision plan, quiz yourself, and check your own answers. Includes reusable prompts and the habits that stop AI from weakening your learning.

Prompts for resumes, LinkedIn and interview practice

Use AI to sharpen your resume bullets, tailor them to a job description, improve your LinkedIn profile and practise interviews, without inventing experience. Includes copyable prompts and the honesty rules recruiters care about.

Prompts for coding: explain, generate, review, debug

Use AI as a coding partner without losing control of your code. Learn four prompt patterns (explain, generate, review, debug) with a runnable Python example for each, and the habit of always running and testing what AI gives you.

Prompts for work: emails, meetings and reports

Use AI to draft clear emails, plan meetings, turn messy notes into minutes with action items, and write status reports, while keeping confidential data safe and every fact checked.

Prompt templates and your personal prompt library

Stop rewriting the same prompts. Turn your best prompts into templates with clear placeholders, store them where you can reach them in seconds, and keep them working as AI tools change.

Testing and evaluating prompts

Learn to test a prompt like an engineer: build a small test set, define good answers with hard and soft checks, score with a short Python script and compare versions so you know a change really helped.

Prompt injection and safe prompting

Learn what prompt injection is, why AI tools can be tricked by text hidden in documents and web pages, and the layered habits that reduce the risk, plus everyday rules for not leaking private data into AI tools.

Project: build a prompt toolkit for your team

Build a shared library of tested prompt templates and a small Python tool that lists, fills and checks them. Practise templates, testing and safe handling of untrusted text in one project a team can really use.