Retrieval & Agents
Agent Memory
The mechanisms that let an AI agent remember things across steps or sessions, beyond what fits in one context window.
Because the context window is finite and resets, memory is what gives an agent continuity. Short-term memory lives in the current context as the running conversation and scratchpad, while long-term memory is stored outside, often in a vector database, and retrieved when relevant. Good memory lets an agent recall earlier decisions, user preferences, and facts it learned, which is essential for multi-step and long-running tasks.
Related terms
AI Agent
An AI system that can take actions and use tools to pursue a goal, not just answer a single question.
Context Window
The maximum amount of text, measured in tokens, that a model can consider at once.
Vector Database
A database built to store embeddings and quickly find the items most similar in meaning to a query.
Retrieval-Augmented Generation
A technique that looks up relevant documents and feeds them to a model so its answers are grounded in real sources.