AI Support & Agents

Agent Memory

Definition

Agent memory is an AI agent's ability to retain and recall information across a conversation — or across sessions — such as earlier messages, customer details, and past interactions, so it stays coherent instead of treating every turn as brand new.

Also known as: conversational memoryAI memorycontext retention

How it works

Memory typically comes in layers. Short-term memory holds the current conversation within the model's context window. Long-term memory stores information beyond a single session — often in a database or vector store — and retrieves the relevant parts when needed, so the agent can remember a customer's prior issues or preferences later.

Because a model's context window is finite, longer histories are usually summarized or selectively retrieved rather than fed in wholesale, keeping the most relevant details available without overflowing the window.

Why it matters for support

Memory is what makes an AI agent feel like a continuous assistant rather than a goldfish. It lets the agent reference what was already said, avoid asking customers to repeat themselves, and carry context across channels and visits — a major driver of a smooth experience.

Frequently asked

What is the difference between short-term and long-term agent memory?

Short-term memory is the current conversation held in the context window; long-term memory persists across sessions in external storage and is retrieved when relevant.

Does agent memory use the context window?

Short-term memory does. Long-term memory lives outside the window and is selectively pulled in, since the window is finite and can't hold unlimited history.

Related terms

Context Window

The context window is the maximum amount of text — measured in tokens — that a language model can consider at once, including both the input you give it and the output it generates..

Vector Database

A vector database is a database built to store and search embeddings — high-dimensional vectors — by similarity, so you can quickly find the content whose meaning is closest to a query..

Embeddings

Embeddings are numeric vector representations of text (or images, code, etc.) that capture meaning, so that pieces of content with similar meaning sit close together in a high-dimensional space..

AI Agent

An AI agent is a software system that uses a language model to understand a request, decide what to do, and take actions — calling tools, retrieving data, and completing tasks — rather than just returning a single scripted answer..

Conversational AI

Conversational AI is technology that lets people interact with software through natural back-and-forth dialogue — understanding intent, maintaining context across turns, and responding in natural language via chat or voice..

Put these ideas to work

Macha is an AI agent layer that sits on top of the help desk you already run — Zendesk, Freshdesk, Front, Intercom, or Gorgias.

Start Trial