Agency17 June 20264 min read
A five-layer memory for an AI agent that works a client for nine months
A context window is not memory. For an agent that handles one client's accounting across nine months, we built memory as five distinct layers (facts, history, decisions) with a promotion path that turns a one-off ruling into a standing rule.
The short answer
A context window is not memory. For an AI agent that handles one client's accounting over nine months, we built memory as five distinct layers: a graph of mandates and rules, a fast working store, a decision journal, a semantic archive of past communication, and an activity log. A promotion path moves a repeated decision up into a standing rule. Different kinds of remembering need different stores, and collapsing them into "stuff the context window" is why long-running agents forget what matters.
Short version: a context window is not memory. For an agent that handles one client's accounting across nine months, the hard part was not the model, it was remembering: the rules of this specific mandate, the decisions made along the way, the history of what was said and done. We built that as five distinct layers, each answering a different question, with a promotion path that turns a repeated decision into a standing rule. Collapse all of that into "stuff more into the context window" and the agent forgets exactly the things a long client relationship depends on.