Context Engineering
ELI5 — The Vibe Check
The art of feeding the right information to an AI so it gives better answers. It's not just about what you ask (prompt engineering) — it's about what background info, examples, and constraints you provide. Like briefing a new hire: the better the briefing, the better the work.
Real Talk
Context engineering is the practice of carefully curating and structuring the information provided to an LLM to maximize output quality. It encompasses prompt engineering, retrieval-augmented generation, system prompt design, few-shot examples, and tools like CLAUDE.md or .cursorrules that provide project-specific context.
Show Me The Code
// CLAUDE.md — context engineering in action
# Project Context
- Stack: Nuxt 4, TypeScript, Supabase
- Style: Composition API, <script setup>
- Rules: No any types, always use Danish for UI text
When You'll Hear This
"Prompt engineering is yesterday's news — context engineering is what matters." / "Your CLAUDE.md file IS context engineering."
Related Terms
CLAUDE.md
CLAUDE.md is a special file you put in your project to give Claude Code persistent instructions — your project's personality file.
Prompt Engineering
Prompt engineering is the art of talking to AI so it actually does what you want.
RAG (Retrieval Augmented Generation)
RAG is how you give an AI access to your private documents without retraining it.
System Prompt
A system prompt is the secret instruction manual you give the AI before the conversation starts. It sets the personality, rules, knowledge, and behavior.