Terms & explanations
Our knowledge base with plain-language explanations of common terms from software development and IT.
Hallucination
A hallucination is an invented but fluently written answer from a language model. It follows from how the model works and cannot be trained away, only contained.
Human-in-the-loop
Human-in-the-loop means a person decides, approves or corrects at a defined point in an AI system's workflow, rather than wherever it happens to be convenient.
Large Language Model (LLM)
A large language model is trained on vast amounts of text to predict the next fragment. Answers, summaries and translations all emerge from that single capability.
llms.txt
llms.txt is a text file at a website's root that tells AI systems, compactly, what content exists and where the authoritative versions live.
Model Context Protocol (MCP)
MCP is an open standard for how AI applications reach tools and data sources. Build an integration once and any application that speaks the standard can use it.
On-premise AI
On-premise AI means running models on infrastructure you control rather than a vendor's. Your data does not leave your own environment to be processed.
Retrieval Augmented Generation (RAG)
RAG connects a language model to a search over your own documents. The model answers from the retrieved passages instead of relying on what it learned in training.
Structured data
Structured data is machine-readable markup in a page's source explaining what the page says. It attaches statements to an entity.
Vector database
A vector database stores embeddings and finds the entries closest in meaning to a query. It is the search engine behind most AI applications that work with your own data.
Zero-click search
A zero-click search ends without a click on any result because the answer sits in the search itself. For websites that means visibility without a visit.
What work could agents take off your team?
Bring one recurring workflow or a product idea. Thirty minutes, an honest assessment, a clear next step.