██████╗ ██████╗ ████████╗ █████╗
██╔══██╗██╔═══██╗╚══██╔══╝██╔══██╗
██████╔╝██║ ██║ ██║ ███████║
██╔══██╗██║ ██║ ██║ ██╔══██║
██║ ██║╚██████╔╝ ██║ ██║ ██║
╚═╝ ╚═╝ ╚═════╝ ╚═╝ ╚═╝ ╚═╝
Core Mission • System Ecosystem • Technical Architecture • Engineering Principles
ROTA is dedicated to revolutionizing the educational landscape by integrating advanced large language models with high-fidelity MetaHuman avatars. Our objective is to build a stateless, high-performance orchestration layer that adapts dynamically to individual cognitive profiles, linguistic patterns, and learning velocities.
We believe that true education is not the delivery of content, but the orchestration of an intellectual spark.
The ROTA architecture is implemented as a decoupled polyrepo system, ensuring independent scalability and specialized domain focus across the stack.
| Repository | Focus Area | Status |
|---|---|---|
| rota-platform | Centralized application orchestrator and user lifecycle management. | Active |
| rota-ai-engine | Single Source of Truth for Gemini Live API integration and tool schemas. | Stable |
| rota-course-engine | Worker-based pipeline for automated syllabus and content generation. | Scaling |
| rota-atlas | Real-time knowledge graph visualization and navigation system. | Alpha |
| rota-canvas-engine | Low-latency rendering engine for interactive educational diagrams. | Development |
Fig 1. Logical Architecture Schematic
The platform leverages a hybrid cloud-client execution model:
- Orchestration: Direct integration with Gemini 2.5 Flash and Pro via standard REST and Realtime WebSocket protocols.
- Rendering: SvelteKit-powered frontend utilizing hardware-accelerated HTML5 Canvas and WebGL.
- Synchronization: Event-driven architecture ensuring eventual consistency across distributed engines.
- Avatars: Future integration with Unreal Engine 5 Pixel Streaming for photorealistic pedagogical interaction (Phase 6).
- Single Source of Truth (SSOT): Critical model identifiers and prompts are managed centrally to prevent logic drift.
- Dependency Injection: Systems are designed to accept dependencies at runtime, facilitating robust unit testing and mocking.
- Fail Fast Execution: Strict input validation and early exit patterns are enforced across all services.
- Domain Decoupling: Business logic is isolated from infrastructure, ensuring that engines remain framework-agnostic.
ROTA Education | Istanbul & Remote Engineering