Glossary
What is context engineering?
Choosing and shaping what an AI model sees, rather than rewriting the question.
Prompt engineering tunes the instruction. Context engineering tunes the evidence: which documents are included, in what order, with what labels.
In practice the biggest wins come from removing irrelevant material and labelling each source clearly so the model can attribute claims.
It matters most for multi-document questions, where the model has to compare rather than recall.
In practice
Numbering each page and prefixing it with its title and URL lets the assistant cite source three by name instead of saying 'one of the pages'.
Common questions
- What does context engineering mean in one sentence?
- Choosing and shaping what an AI model sees, rather than rewriting the question.
- Why does context engineering matter when you use Claude or ChatGPT?
- It matters most for multi-document questions, where the model has to compare rather than recall.
- Can you give an example of context engineering?
- Numbering each page and prefixing it with its title and URL lets the assistant cite source three by name instead of saying 'one of the pages'.
Related terms
- Context windowThe amount of text an AI assistant can hold in mind during a single conversation.
- GroundingTying an AI answer to specific source material you supplied.
- Prompt engineeringWriting the instruction so the model does what you meant.
- Glossary indexEvery AI context term, defined in plain language.
Get LocalBridge
One click sends your open tabs into Claude or ChatGPT. One-time licence, no subscription.
See how it works