What is an AI skill?
The know-how an AI agent loads only when a task calls for it, and why that makes a skill different from a tool.
What is an LLM (large language model)?
Behind every answer is one repeated act, predicting the next token, and almost everything else about a language model follows from it.
What is context engineering?
Context engineering is the discipline of deciding what fills a language model's context window at inference time: retrieval, history compression, tool schemas, and memory together.
What is fine-tuning?
How a pre-trained model shifts to a specific task or style, and why it is not the right tool for adding new knowledge.
What is MCP (Model Context Protocol)?
One open standard for wiring an AI application to the outside world, whatever model runs inside it.
What is multimodal AI?
How systems that combine text, images, audio, and video in a single model work, and where the modality gap still matters.
What is prompt engineering?
The craft of writing and structuring the text sent to a model, and why each choice measurably changes what comes out.
What is RAG (retrieval-augmented generation)?
How a frozen language model answers from knowledge it was never trained on, by retrieving passages at the moment it responds.
Nothing here yet under that search.
Stop reading about it. Start directing it.
Briefly explains the ideas. Brief is the team that runs on them — opening to a small group at a time.