In the current state of agentic memory most of the context is stored in the form of a MD file or is derived through a RAG model where you store each and everything. Both of the solution leads to bloated context which does not optimize the usage of any tokens.
In this system we only keep relevant data in our memory and prune all the unnecessary data. The relevance of a data is derived through multiple factors such as recall rate, importance, category, to which memory chain it's connected to etc. These parameters are fine tuned so that we can cater to both episodic memory and semantic memory.
Our memory layer keeps the size flat in this manner. You can draw correlation of this infrastructure with how Human brain store and prune memory.
The enterprise model is something very exciting as we can extract relevant memories from each user, agent and sub agent in this layer and that can be used by any one in the org, ensuring memory optimization at an enterprise level.