AI

Context engineering is the more precise term for what people used to lump under prompt engineering: deliberately deciding what goes into a model's context window across an entire application, not just crafting one clever prompt. This course covers why context limits are a real, permanent engineering constraint even as windows get bigger, how to manage conversation history, retrieval, and tool definitions so they don't crowd each other out, how prompt caching fits into the picture, and why treating context as a scarce resource produces applications that are both cheaper and noticeably better.