AI & Automation Glossary

An autonomous software system powered by large language models that can reason, plan, and take actions to accomplish tasks with minimal human intervention.

The orchestration framework that manages an AI agent's lifecycle, including tool access, memory, guardrails, and execution flow within a production environment.

Dense numerical vector representations of text, images, or other data that capture semantic meaning, enabling machines to measure similarity and perform search.

The process of further training a pre-trained language model on a domain-specific dataset to improve its performance on specialized tasks.

A class of artificial intelligence systems that create new content — text, images, code, or structured data — based on patterns learned from training data.

Safety mechanisms and constraints applied to AI systems to prevent harmful outputs, enforce compliance policies, and keep agent behavior within defined boundaries.

When an AI model generates plausible-sounding but factually incorrect or fabricated information, often occurring when the model lacks sufficient context or training data.

A deep learning model trained on massive text datasets that can understand, generate, and reason about natural language. Examples include GPT-4, Claude, and Llama.

The practice of designing and refining input prompts to elicit accurate, relevant, and useful responses from large language models.

An architecture that enhances LLM responses by first retrieving relevant documents from a knowledge base, then using them as context for generation — reducing hallucinations and grounding outputs in real data.

An AI agent's ability to invoke external tools — APIs, databases, code interpreters, or web browsers — to gather information or take actions beyond its built-in capabilities.

A specialized database that stores and indexes high-dimensional vector embeddings, enabling fast similarity search across documents, images, or other unstructured data.