The Personal Assistant We'll Build › Glossary

AI Agent: A program powered by a language model that can perceive its environment, make decisions, and take actions autonomously to achieve goals. Unlike simple chatbots, agents can use tools, remember context, and plan multi-step solutions.

Language Model: An AI system trained on vast amounts of text that can understand and generate human-like language. In an AI agent, the language model processes requests and formulates responses.

Prompt: The instructions and context you provide to a language model to guide its behavior and responses. Think of it as how you communicate your needs to the AI.

Tool Use: The ability of an AI agent to call external functions or APIs to perform tasks beyond text generation, such as calculations, web searches, or database queries.

Memory: The agent's ability to retain information across interactions, including both short-term context (recent conversation) and long-term knowledge (stored facts and preferences).

State: The collection of information an agent maintains about its current situation, including the user's goal, conversation history, intermediate results, and available tools.

Planning: The process by which an agent breaks down complex tasks into smaller, manageable steps and executes them in a logical sequence to achieve a goal.

Probabilistic: A characteristic of AI systems where outputs are based on calculated likelihoods rather than deterministic rules, meaning the same input may produce slightly different outputs across runs.