Deconstructing Project89: A modular, high-performance new AI Agent framework

Project89 has proposed a brand new Agent Framework design, primarily aimed at game development, featuring a higher degree of modularity and performance. This framework adopts the ECS(Entity-Component-System) architecture, thoroughly separating data from logic to achieve efficient management in large-scale scalable scenarios.

Core Architecture of the Framework

The core architecture of Project89 mainly includes the following parts:

1. Entity(: exists only as an ID, containing no data or logic.

2. Component) component(: Used to store specific data or state of entities. It can be divided into core identity types, behavior and state types, perception and memory types, environment and space types, appearance and interaction types, auxiliary or operation and maintenance types, etc.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-4cd7ca20f2967b9025411d9985f64831.webp(

3. System (System ): Responsible for executing logic related to specific components. The system is divided into three levels:
   • Consciousness Systems: such as PerceptionSystem, ThinkingSystem, etc., with a relatively high update frequency.
   • Subconscious System: such as GoalPlanningSystem, the update frequency is relatively low.
   • Unconscious System: Currently not enabled, minimum update frequency.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-69e16ae3b5af8c39e1fc522358584cef.webp(

4. Manager) Manager(: Provides system-level functions such as RoomManager, StateManager, etc., responsible for resource management and data access.

5. SimulationRuntime: As the scheduler for all Systems, responsible for starting or stopping the system loops at different levels.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-59f1984837d6636d7efc378c731a01eb.webp(

Innovations and Advantages

1. Highly modular: Each System operates independently, with no inter-call relationships, making it easy to expand and customize.

![Deconstructing Project89: A Modular, High-Performance Next-Gen AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-c37bd7442c3743f80a4fd9ba8afc123b.webp(

2. High Flexibility: The capabilities of the Entity can be easily increased or decreased according to needs, simply by adjusting the registered Components and Systems.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-19636582e09b473536b17c2de0c61fbc.webp(

3. High Performance: The ECS architecture is more suitable for concurrent processing, showing excellent performance in complex scenarios.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-b0180b99743a98bafb2b2d066900d65c.webp(

4. Layered Design: Divide the System into three layers: conscious, subconscious, and unconscious, simulating the human cognitive process to achieve more realistic AI behavior.

![Deconstructing Project89: A Modular, High-Performance Next-Gen AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-6a3ba796aae66b8a9f977ec99b5bb1c8.webp(

5. Data Management: Achieve interaction between ECS and the database through StateManager, supporting state persistence and recovery.

![Deconstructing Project89: A Modular, High-Performance Next-Generation AI Agent Framework Design])https://img-cdn.gateio.im/webp-social/moments-aadac69287c878cdd84a72c6b8929ef2.webp(

The design of Project89 is not only suitable for game development but also provides new architectural options for other complex AI applications like quantitative trading ). Its highly modular structure, excellent performance, and clear design philosophy offer a highly promising solution for developing large-scale, complex AI systems.