Learn Prompt Engineering
How to write what goes in the box, so the answer that comes back is the one you wanted. Sixteen lessons start with what a prompt is, how a chat works and what a system prompt does. Then come tags, examples, steps and chains. After that, prompts inside an app: set shapes, caching and picking a model. The last four lessons cover prompts for code. It assumes nothing. It stops at the prompt. Training a model is a separate subject.
Not started
0 of 16 lessons done
Level 1 · What a prompt is
score0250/11 wordsThe text you send and the tokens it is counted in. How a chat is sent again each time, and the rules that sit above it.
- Prompt Engineering BasicsWhat a prompt is and what prompt engineering adds. The three jobs a good prompt does, and why one good answer proves very little.10 min4 terms
- Multi-Turn ConversationsWhy a chat model seems to remember you, and what it costs. When to start a fresh chat, and when the regenerate button is the right fix.10 min3 terms
- System Prompts and RolesWhat a system prompt is, the three message roles, and the custom instructions you save yourself. What a role changes, and what it never adds.12 min4 terms
Level 2 · Writing a better prompt
score25400/7 wordsFencing off the parts of a prompt with tags. Showing the model worked examples, then fixing the prompt one change at a time.
- Prompt Delimiters and TagsHow tags and other delimiters keep your instructions apart from the text the model works on. Where to put each part, and what a tag cannot stop.10 min2 terms
- Zero-Shot and Few-Shot PromptsThe difference between zero-shot, one-shot and few-shot prompts. When a worked example helps, how to pick good ones, and what they cost.12 min3 terms
- Iterative and Meta-PromptingHow to improve an answer one change at a time, and how to ask a model to write the prompt for you. Why both still need a test on real cases.10 min2 terms
Level 3 · Harder and bigger tasks
score40650/9 wordsAsking for the working before the answer. Splitting a big job into a chain of prompts, and choosing what the model gets to see.
- Chain-of-Thought PromptingWhy asking for the steps makes hard answers better, with the real numbers. When to vote across runs or branch, and when it is all a waste.12 min4 terms
- Prompt Chains and TemplatesHow to split a big job into a chain of short prompts. How a template with blanks lets code send the same tested prompt each time.12 min2 terms
- Context EngineeringChoosing everything a model sees for one request, not just the prompt. Why more text can make answers worse, and how to pick, order and test it.12 min3 terms
Level 4 · Prompts inside an app
score65850/10 wordsHow to make replies a program can read. How to pay less for the same prompt, and how to test models on your own examples.
- Structured Output and JSONHow structured output makes a model's reply fit a JSON schema. How it differs from JSON mode, and why a reply that fits can still be wrong.12 min4 terms
- Prompt Caching and CostHow prompt caching reuses the long fixed start of a prompt for far less. Why the order of a prompt now matters, and where else the cost hides.11 min2 terms
- Model Selection and TestingHow to pick the AI model that runs a prompt. Build a small test set from real examples, then choose the smallest model that passes it.12 min4 terms
Level 5 · Prompts for code
score851000/11 wordsAsking an AI for code and giving a task to an agent. Writing the plan first, then checking what comes back before it piles up.
- Code Generation and CompletionHow AI code completion and code generation are not the same. How to prompt each one, and why code that runs on the first try still has to be read.11 min3 terms
- Agentic Coding and AGENTS.mdHow a coding agent works through a task on its own. How to write a task it can finish, and what goes in the AGENTS.md file it reads first.12 min3 terms
- Vibe Coding vs Spec-DrivenWhat vibe coding is, what spec-driven development is, and when each fits. Why the real mistake is shipping the quick version as if it were the real one.11 min2 terms
- AI Code Review and Tech DebtHow AI code review helps and where it stops, and how AI-written code turns into debt. The old habits that keep a fast team from slowing down.11 min3 terms
Words in this course
The 48 glossary entries this course teaches, in the order it teaches them. Open any of them on its own.
- Prompt (AI)
- Prompt Engineering
- Large Language Model (LLM)
- Token (AI)
- Multi-Turn Conversation
- Context Window
- Regenerate Response
- System Prompt
- Message Roles
- Custom Instructions
- Role Prompting
- Prompt Delimiter
- Prompt Injection
- Zero-Shot Prompting
- One-Shot Prompting
- Few-Shot Prompting
- Iterative Prompting
- Meta-Prompting
- Chain-of-Thought Prompting
- Self-Consistency (Prompting)
- Tree of Thoughts (ToT)
- Reasoning Model
- Prompt Chaining
- Prompt Template
- Context Engineering
- Lost in the Middle
- Context Rot
- Structured Output (LLM)
- JSON Mode
- JavaScript Object Notation (JSON)
- JSON Schema
- Prompt Caching
- Max Tokens
- Model Selection (LLM)
- Frontier Model
- Small Language Model (SLM)
- LLM Evaluation
- AI Code Completion
- Code Generation (AI)
- AI Coding Assistant
- Agentic Coding
- Agent Instruction File
- AI Agent
- Vibe Coding
- Spec-Driven Development
- AI Code Review
- AI Technical Debt
- Code Review