The words enterprise AI is sold with, defined honestly.
Written by the engineers who build with these things, not by a marketing team. Where a technique is usually the wrong choice, we say so — 12 terms, no hype.
AgentOps
AgentOps is the practice of operating AI agents in production — enforcing what an agent may do, monitoring what it costs, and recording why it acted.
Agentic AI
Agentic AI describes systems that pursue a goal across multiple steps — planning, calling tools and completing work — rather than returning a single response.
Graph-RAG
Graph-RAG is retrieval-augmented generation that retrieves over a knowledge graph, so answers can follow explicit relationships between entities instead of relying on text similarity.
Retrieval-Augmented Generation (RAG)
Retrieval-augmented generation is a technique where a language model retrieves relevant passages from a specific body of data and answers using them, rather than from its training alone.
Model Context Protocol (MCP)
The Model Context Protocol is an open standard for exposing tools, data and actions to AI models through a consistent interface, so an agent can call them without bespoke integration each time.
AI-native retrofit
An AI-native retrofit adds agentic intelligence to software that is already running, by layering over its existing interfaces rather than rebuilding the system.
Governor Engine
A Governor Engine is a runtime control layer that decides whether an AI agent's proposed action is permitted, within budget and safe to execute before it happens.
Human-in-the-loop
Human-in-the-loop describes an AI system designed so that a person reviews or approves defined decisions rather than the system acting entirely unattended.
Multi-agent orchestration
Multi-agent orchestration is the coordination of several specialised AI agents on one task, each handling part of it under a supervisor that can halt the whole run.
MLOps
MLOps is the discipline of deploying, versioning, monitoring and retraining machine learning models in production reliably.
Fine-tuning
Fine-tuning is further training an existing model on a specific dataset so it performs better on a narrower task or domain.
Contract intelligence
Contract intelligence is the use of AI to extract and reason over the terms, obligations and relationships inside contracts, with each answer traceable to its source clause.