Prompting Foundations
The bedrock skills of prompt engineering — how models read your words, the anatomy of a strong prompt, and how to steer tone, format, and accuracy.
What a prompt really is
A prompt is an instruction, not a search query When you use a search engine, you type keywords and it finds pages that contain them. A language model works differently: it reads your whole prompt as instructions and writes a fresh answer, word by word, that it predicts should follow. That single difference changes everything. […]
Role, context, task, format
Four parts that turn a request into a brief Almost every strong prompt contains four ingredients. You do not need fancy wording — just make sure each is present. 1. Role Tell the model who to be. “You are a senior financial analyst” primes it toward the right vocabulary, assumptions, and level of rigour. 2. […]
Where models are strong and where they are not
Knowing when not to ask is part of the skill A language model predicts fluent text. Everything it is good at, and everything it is bad at, follows from that one fact — and fluency is not the same thing as being right. Reliably strong Transforming text you supply — summarising, rewriting, translating, changing register, […]
Build your own prompt checklist
Turning this course into something you actually use Technique you have to remember is technique you will skip when you are busy. The fix is a short checklist you keep beside you until it becomes automatic — the same reason surgeons and pilots use them, and for the same reason: not because the steps are […]
Choosing the cheapest pattern that works
Every pattern costs something Patterns are usually taught as pure upside. In production they are trade-offs: examples cost input tokens on every single call, reasoning costs output tokens and latency, chains cost round trips, self-checking multiplies everything by two or three. The ladder Start at the bottom and stop as soon as the output is […]