Too little context and AI guesses; too much and it is distracted; the wrong context and it is confidently wrong. Give AI the context it needs.
Context engineering
The practice of deciding what information, tools, instructions and history an AI receives for a task, and in what form.
Example
For a client query, the AI receives that client’s current status, recent correspondence and the relevant Skill, but not other clients' files or the firm’s internal risk notes.