ET is a C# framework for building Unity3D clients and game servers designed specifically to support AI-assisted development of large-scale online games.
The framework addresses the challenge of making AI code generation practical for complex multiplayer games. Rather than treating AI as a simple code generator, ET provides a complete development pipeline where AI-generated code must compile successfully, pass automated gameplay tests, obey architectural rules, and survive hot reload cycles before reaching production. It enforces these constraints through a powerful analyzer ecosystem that validates async safety, entity lifecycles, package dependencies, actor usage, and runtime safety, allowing AI to work within strict architectural boundaries instead of producing uncontrolled code.
ET suits teams building distributed MMO architectures who want to leverage AI coding tools without sacrificing project structure or stability. The framework is explicitly designed for large-scale online games rather than standalone projects, providing a modern actor-based runtime with dynamic room scaling, service discovery, and cloud-native deployment. It includes direct Unity Editor integration that lets AI interact with the engine runtime to create GameObjects, modify components, control Play Mode, and run gameplay tests, enabling true AI-driven workflows beyond file-based code generation. Teams should adopt this if they need both the scalability of a distributed server architecture and the development velocity that AI assistance can provide, but not if they are building small or single-player projects.
The project maintains active development with regular commits across its distributed architecture components, analyzer systems, and Unity integration layers. Work spans multiple technical domains including the fiber-based actor runtime, cloud-native server infrastructure, and the AI harness verification pipeline, indicating sustained effort to keep the framework's core systems current. The codebase shows ongoing refinement of the analyzer ecosystem and hot reload workflows that form the foundation of the AI-assisted development model.
