TL;DR

• •Enterprise AI agents frequently fail to move beyond pilot phases due to their inability to remember validated actions and apply explicit decision context.
• •Traditional RAG (Retrieval Augmented Generation) is excellent for surfacing documents but falls short for agents requiring time-aware reasoning and non-regressive, actionable intelligence.
• •Decision context graphs provide structured memory and explicit logic, allowing agents to understand what rules apply, when they apply, and to build reliably on past decisions.

TL;DR

• •YourMemory implements an AI memory system using the Ebbinghaus forgetting curve.
• •It demonstrates a 52% recall rate, outperforming Mem0 on the LoCoMo benchmark by 16 percentage points.
• •The project is open source and available on GitHub for developers to experiment with.

TL;DR

• •Enterprise AI agents frequently fail to move beyond pilot phases due to their inability to remember validated actions and apply explicit decision context.
• •Traditional RAG (Retrieval Augmented Generation) is excellent for surfacing documents but falls short for agents requiring time-aware reasoning and non-regressive, actionable intelligence.
• •Decision context graphs provide structured memory and explicit logic, allowing agents to understand what rules apply, when they apply, and to build reliably on past decisions.

TL;DR

• •YourMemory implements an AI memory system using the Ebbinghaus forgetting curve.
• •It demonstrates a 52% recall rate, outperforming Mem0 on the LoCoMo benchmark by 16 percentage points.
• •The project is open source and available on GitHub for developers to experiment with.