Glossary
The words, in plain English
What the terms behind AI operations actually mean, and how they work in Oido. Written for the person deciding, not the person building.
Agents
AI agentAn AI agent is software that pursues a goal on its own, deciding steps, using tools and acting without a human driving each move. The short definition.AI copilotAn AI copilot assists a person in real time with suggestions and drafts, while the human stays in control. How it differs from an agent.AI employeeAn AI employee is an AI agent set up to own a recurring role, handling tasks end to end across your tools and channels, not just answering questions.multi-agent orchestrationMulti-agent orchestration coordinates several specialized AI agents on one goal, with an orchestrator that delegates to subagents.system promptA system prompt is the instruction that sets an AI agent's role, rules and behavior before any conversation starts. How to write one.tool callingTool calling is how an AI agent uses external functions, searching, reading a database, sending a message, instead of only writing text.
Automation
agentic automationAgentic automation uses AI agents that decide how to complete a task, not just follow fixed rules. How it differs from workflow automation.no-code automationNo-code automation lets non-developers connect apps and automate work using visual tools instead of code. What it is, and its limits.workflow automationWorkflow automation runs a fixed sequence of steps across your apps automatically. How it works, and how it differs from AI agents.
Operations
Standards
Model Context Protocol (MCP)The Model Context Protocol (MCP) is an open standard that lets AI agents connect to external tools and data through one interface.multi-provider AIMulti-provider AI means your agents can run on any model provider, Claude, GPT, DeepSeek and more, instead of being tied to one vendor.
Foundations
large language model (LLM)A large language model (LLM) is an AI trained on huge amounts of text to understand and generate language. What LLMs power, and their limits.RAG (retrieval-augmented generation)Retrieval-augmented generation (RAG) lets an AI look up relevant information before answering, grounding its response in your data.
Deployment