AI Dictionary
Plain-language explanations of AI and automation terms – no jargon, just what you actually need to know.
AI Agent
Software that receives a task, plans how to complete it, executes each step, and reports the result – without you managing every move.
Read more →LLM (Large Language Model)
The brain behind ChatGPT, Claude, and Gemini. A model trained on billions of texts that understands language and communicates fluently.
Read more →RAG – Retrieval-Augmented Generation
A technique that lets AI "read" your company documents and answer based on them – not just from its training data.
Read more →Prompt Engineering
The art of formulating instructions for AI so you get exactly what you need – not something that just looks right.
Read more →Process Automation
Setting up a system that executes repetitive steps for you – without manual intervention every time.
Read more →MCP – Model Context Protocol
The standard "language" AI models use to communicate with external tools and data sources.
Read more →Fine-tuning
Adapting an existing AI model for a specific task or style using your own data.
Read more →Vector Database
A special type of database that stores texts as mathematical patterns – letting AI find "similar" documents, not just exact matches.
Read more →Chatbot vs. AI Agent
A chatbot answers pre-scripted questions. An AI agent understands context, makes decisions, and acts – even in situations it wasn't explicitly programmed for.
Read more →n8n (Workflow Automation)
An open-source tool for visually connecting applications and automating processes – without writing code.
Read more →API (System Integration)
The way different software systems "talk" to each other – CRM, email tools, invoicing software, AI agents.
Read more →AI Hallucination
When AI confidently states something that isn't true. Not intentional – a known property of language models that needs to be managed.
Read more →Context Window
The maximum amount of text an AI model can "see" at once – its working memory for a single conversation.
Read more →Token
The basic unit AI models use to process text – roughly a syllable to a word. Also the basis for API billing.
Read more →Embedding
Converting text into numbers so that similar texts have similar numbers – the foundation for semantic search and RAG systems.
Read more →Knowledge Base
An organised collection of company information, documents, and rules that AI agents draw from when answering questions.
Read more →RPA – Robotic Process Automation
Software that controls a computer like a human – clicking, copying, filling forms – without needing an API.
Read more →No-code / Low-code Automation
Tools that let you build automations and applications without writing code – or with very little of it.
Read more →