Glossary

AI memory, defined

Short, plain definitions of the terms behind shared memory for AI-native teams.

Company Brain

A Company Brain is a context layer that captures, structures and retrieves what a company has discussed, decided and learned across people, AI agents and documents.

See full definition

AI-agent conversation

An AI-agent conversation is a working session between a person and an AI agent, such as a Claude Code session.

See full definition

AI memory layer

An AI memory layer is an external system that captures information from AI interactions and recalls it in later sessions, giving otherwise stateless models persistent context across time.

See full definition

Shared context

Shared context is the working knowledge (decisions, conventions, and background) that every AI tool and teammate can read from one place, instead of each keeping its own isolated copy.

See full definition

Agent memory

Agent memory is the ability of an AI agent to store and recall facts, decisions, and outcomes across runs, so it improves over time instead of restarting from scratch each session.

See full definition

Context window

A context window is the amount of text a model can consider at once; anything outside it, including past sessions, is invisible to the model unless re-supplied.

See full definition

Second brain for teams

A second brain for teams is a shared, searchable store of a team's knowledge and decisions that both people and their AI tools can draw on, so institutional memory does not live only in individual heads.

See full definition

Retrieval-augmented generation (RAG)

RAG is a technique where a model retrieves relevant documents from a knowledge base and uses them to ground its answer, rather than relying only on what it was trained on.

See full definition

Session memory

Session memory is context that lives only for the duration of a single AI session and disappears when it ends, unlike persistent memory which survives across sessions.

See full definition

Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open standard that lets AI tools connect to external systems, such as a memory layer, through a common interface.

See full definition