Applied-AI glossary

Clear definitions of agentic AI, AI agents, orchestration, RAG and other key concepts for putting AI into production across your organization.

  • What is an AI agent? — An AI agent is a system that perceives its context, decides the next step, uses tools and data, and executes actions to meet a goal with minimal human intervention.
  • AI agent vs chatbot — A chatbot responds within a conversation; an AI agent also decides, uses tools and runs multi-step actions to complete a task.
  • What is an Agentic OS? — An Agentic OS (agentic operating system) is a bespoke orchestration layer that connects an organization's processes, data and tools so AI agents do real work in production, under human control. How it differs from a standalone agent.
  • What is an enterprise copilot? — An enterprise copilot is an AI assistant embedded in a team's tools that suggests, drafts or executes tasks while keeping the person in control of the decision.
  • What are evals for AI agents? — Evals are systematic tests that measure whether an AI agent meets its goal with the required quality and reliability before production: what is measured, how they are designed and how they differ from classic tests.
  • Fine-tuning vs RAG — Fine-tuning retrains a model with your data; RAG gives it access to your data when answering. You choose based on freshness, cost and control of information.
  • What is AI governance? — AI governance is the set of rules, controls and responsibilities that ensure an organization's AI systems are traceable, supervised and auditable.
  • What is human-in-the-loop (HITL) in AI? Where to place human control — Human-in-the-loop (HITL) is a design in which a person reviews, approves or corrects an AI system's decisions at the points where human judgment matters. Patterns, enterprise examples, the difference from human-on-the-loop and what the EU AI Act requires.
  • What is agentic AI? — Agentic AI is an approach in which AI systems plan, decide and execute actions autonomously to reach a goal, instead of only generating text.
  • What is AI in production? — AI in production is an AI system integrated into an organization's real operation —with live data, human control and traceability— and not a prototype or demo.
  • What is responsible AI? — Responsible AI is the design and use of AI systems in a safe, transparent way, aligned with the organization's rules and values.
  • What is an LLM? — An LLM is an AI model trained on large volumes of text to understand and generate language, and the foundation of most agents and copilots.
  • What is LLMOps / MLOps? — LLMOps is the set of practices to deploy, monitor and maintain language models in production reliably.
  • What is MCP (Model Context Protocol)? — The Model Context Protocol (MCP) is an open standard that lets AI models connect in a uniform way to external tools, data and systems.
  • What is agent orchestration? — Agent orchestration is the coordination of several specialized AI agents, directed by an orchestrator, to solve a complex task from end to end.
  • What is an AI orchestrator? — An AI orchestrator is the component that decides which agents act, in what order and how their results are combined within an agentic system. Explained in plain words, with the difference from a standalone agent.
  • What is RAG? — RAG connects a language model to external information sources to ground its answers. Learn how it works, its limits and the controls needed in production.
  • What is the ROI of an AI project? — AI ROI is the net business value an AI system generates against its cost, measured with a metric agreed before starting.
  • What is a multi-agent system? Examples and what it is for — A multi-agent system is a set of AI agents with distinct roles, coordinated by an orchestrator, that execute complete processes. How it works, enterprise examples and when it pays off over a single agent.
  • What is AI traceability and how do you achieve it? — AI traceability is being able to reconstruct which data, steps and decisions produced each result. What the EU AI Act requires, how it is audited and where human oversight fits.