Context Engineering
Definition
Context engineering is the discipline of structuring what an AI system can see: selecting, organising and delivering the right business context to a model at the right moment, so its output reflects how the organisation actually works rather than a plausible guess.
Prompt engineering tunes the question. Context engineering builds the library the answer comes from. It covers how knowledge is chunked and embedded, how retrieval decides relevance, what standing instructions and memory an agent carries, and what each role is permitted to see.
It earns the word engineering because the failure modes are structural, not stylistic. An agent that answers from stale policy, misses the one document that mattered, or leaks context across roles is not badly prompted, it is badly plumbed. Most enterprise AI disappointment traces here, which is why our whitepaper treats context infrastructure as its own stage rather than a detail of development.
Where this shows up
Definitions are the map. See where your own business sits on it: the audit runs free in your browser, in about two minutes.
Start the audit