Agentic AI Glossary

Defines and deploys autonomous AI systems that plan, execute, and coordinate complex workflows across tools and APIs with minimal human intervention.

LLM agents that invoke external functions, APIs, and services during reasoning to retrieve data, modify state, or perform actions beyond text generation.

Purpose-built framework wrapping LLMs with memory, tool registries, permission systems, and execution loops for domain-specific autonomous operation.

Private, self-hosted systems enabling individuals to deploy autonomous agents with memory, tool access, and persistent identity across sessions.

Persistent storage and retrieval systems enabling agents to recall past interactions, learn from outcomes, and maintain context across long-running processes.

The representation of an autonomous process's current status including pending actions, completed steps, data dependencies, and next decision points.

Autonomous or semi-autonomous systems deployed in commercial settings to execute tasks like lead qualification, CRM updates, document processing, and exception handling.

The strategic amplification of human effectiveness through AI systems that remove repetitive work, preserve human judgment for exceptions, and scale individual output.