Context Engineering
Deciding what information to put in front of a model at the right moment: instructions, data, memory, and tools, so it has what it needs and nothing it does not.
It is prompt engineering grown up. For agents, managing context well matters more than any single clever prompt.
Frequently asked questions
How is context engineering different from prompt engineering?
Prompt engineering is about writing one good instruction, while context engineering is about deciding everything the model should see at the right moment: instructions, data, memory, and tools. Think of it as prompt engineering grown up for more complex AI systems.
Why does context engineering matter for AI agents?
Because an agent takes many steps and juggles lots of information, feeding it the right material at each moment matters more than any single clever prompt. Give it what it needs and nothing it does not, and it performs far better.
Can giving the AI too much information hurt its answers?
Yes, which is a core reason context engineering exists. A model can only hold so much at once, and burying the important details in a flood of extra text can actually make the answer worse, so choosing what to include is part of the skill.
New to all this? Start with what an AI agent really is, browse the full glossary, or explore the learning hub.
← Back to the glossary