Prompt Engineering
A Black Girl's Guide to Prompt Engineering
The magic's in your words. The power is in your prompts.
AI is only as good as the instructions you give it. And if the people using it aren't us, how can it really reflect our needs, our culture, or our voice?
This one takes you from curious user to certified prompt engineer. It starts with the anatomy of a prompt, context, instructions, constraints, and the GIGO rule: garbage in, garbage out. A vague ask gets a vague answer, every time.
From there it gets technical without getting dry: temperature as a creativity dial, token budgets, chaining prompts so one answer feeds the next, role assignment, self-consistency checks, and retrieval-augmented prompts that keep a model grounded instead of hallucinating.
Tone gets its own chapter, boardroom to brunch, because a model that sounds like you is a model you'll actually use. It closes on the career side of the skill and on ethics: spotting bias in what comes back, and refusing to ship it forward.
Key takeaways
- Context, instructions, constraints, the three parts of every strong prompt
- GIGO: how to diagnose a weak answer instead of retrying blindly
- Temperature and token budgets, and when to move each dial
- Role assignment and prompt chaining for multi-step work
- Cultural prompting so responses sound like your world, not a template
- Self-consistency and retrieval to cut hallucinations
- Prompt engineering as a paid career path
Who it's for
Anyone using AI tools daily who wants consistent, professional-grade results.
Paired lab
Practise this book's ideas hands-on.
Chapters
- Chapter 1: What's in a Prompt?Think of prompts like a recipe. If you throw in half the ingredients and skip the instructions, what you end up with is… well, a mess.
- Chapter 2: Engineering Powerful Prompts0.2–0.4 for precision. 0.8–1.0 for wild, creative responses. The dial is yours.
- Chapter 3: Shaping AI to Match Your EnergyFrom boardroom to brunch: prompting with style and soul.
- Chapter 4: Mastering Advanced Prompt EngineeringSelf-consistency: ask the same question more than one way and compare, so the model can't flip-flop on you.
- Chapter 5: Next-Level PromptingThis reduces hallucinations by forcing the model to validate its information through self-checks.
- Chapter 6: The Future of Prompt Engineering as a CareerGet that bag. This is a skill people are being paid for right now.
- Chapter 7: Ethics and Bias MitigationAI comes with baggage, it's been trained on data that may not always see or understand us. We're here to change that.