AI glossary
Artificial intelligence terms in plain English. Every definition starts with what matters most.
AI agent
- Software that understands a request, breaks the task into steps and carries them out with the tools it can access, such as email, a CRM or a database. Unlike a chatbot, it finishes the work. Steps with consequences should be approved by a person.
Agentic AI
- AI that completes tasks on its own instead of only answering. An agentic system plans, uses tools, checks the result and continues until the task is done.
Chatbot
- A program that holds a conversation with a user. A modern AI chatbot answers in natural language, ideally from a company’s verified documents. It usually does not act; that is what an AI agent does.
LLM (large language model)
- An AI model trained on large amounts of text that can write, summarise, translate and answer. Examples: GPT, Claude, Gemini or Mistral models. It is the “brain” behind chatbots and AI agents.
RAG
- Retrieval-augmented generation. The AI first retrieves relevant passages from your documents and answers from them, ideally citing the source. It reduces made-up answers.
Multi-agent system
- Several AI agents, each with a clearly defined role: one reads documents, another checks the data, a third drafts the reply. A coordinating agent or rules keep them in step, and a person approves the important steps.
Process automation
- Replacing repetitive manual work with rules that run on their own: moving data between systems, notifications, approvals, reports. When the input has no fixed shape, AI is added.
Prompt
- The input for an AI model: a question, instruction or context. Output quality depends heavily on how the prompt is written.
Hallucination
- When an AI model confidently states something false. RAG, limiting the model to verified sources and human review all help against it.
AI Act
- The EU regulation on artificial intelligence. It classifies AI systems by risk and sets duties for providers and users. Since 2 February 2025 it includes the duty to ensure AI literacy among staff.
MCP (Model Context Protocol)
- An open standard that lets AI models and agents connect to tools and data, such as databases, files or business apps, in a consistent way.
Bring us one process or one app idea.
In 30 minutes we tell you whether it is worth handing to an AI agent. Then you get a free working prototype within 3 working days. If you don’t like it, you don’t pay.
- We reply within 24 hours on working days
- The engineer who designs it also builds it
- Write in Czech, English, German, Italian or Turkish
Contact
Write to us
Describe the process or app you want done. We reply within 24 hours on working days, and the prototype within 3 working days is free.