Context Engineering

In one sentence

The discipline of designing everything an AI model sees — system prompts, retrieved documents, tool definitions, conversation history, and examples — to produce reliable, high-quality outputs.

In context

A developer building a customer support bot doesn't just write a single prompt. They design a system prompt defining the bot's role and rules, connect a RAG pipeline that retrieves relevant FAQ entries for each question, define tools the bot can call (like order lookup), maintain conversation summaries, and include example responses that show the right tone. Designing how all these pieces work together — what gets included, how much space each piece gets, and what happens when the context window fills up — is context engineering.

See also