Reference · 18 terms
Short, plain-English definitions for the terms AI tools keep using. Each term links to tools that do the thing.
All terms
AI agentSoftware that pursues a goal across multiple steps: it plans, uses tools such as search…Large language model (LLM)A neural network trained on vast text to predict and generate language. LLMs power…RAG (retrieval-augmented generation)A pattern where an AI first retrieves relevant documents and then writes an answer…Prompt engineeringThe practice of writing instructions that get reliable output from an AI: clear goals,…EmbeddingsNumeric representations of text (or images) where similar meanings sit close together.…Fine-tuningFurther training a base model on targeted examples so it adopts a style, domain, or task.…MCP (Model Context Protocol)An open protocol that lets AI assistants connect to external tools and data sources…Agent skillA reusable package of instructions and workflows that gives an AI agent a repeatable…Computer visionAI that interprets images and video: detecting objects, reading text, describing scenes.…Text to speechTechnology that converts written text into spoken audio. Modern neural voices handle tone…Speech to textTechnology that transcribes spoken audio into text. Used for meeting notes, captions, and…Diffusion modelsGenerative models that create images (and video) by iteratively removing noise, guided by…TransformerThe neural architecture behind modern LLMs: attention mechanisms weigh which parts of the…AI chatbotA conversational interface that answers questions in dialogue form. Modern chatbots use…AI copilotAn assistant embedded inside a workflow — an editor, inbox, or IDE — that suggests the…Workflow automationConnecting apps so work moves without manual steps: triggers, conditions, and actions. AI…Vector databaseA database built to store embeddings and find nearest neighbors fast. The standard memory…Multimodal AIModels that accept and combine more than one input type — text, images, audio, video.…