Core Architecture

LuxurAI is designed for maximum speed, security, and developer productivity. By combining a robust Python backend with a multi-layered verification system, it delivers premium AI capabilities seamlessly.

---

Technical Stack Overview

graph LR
    subgraph Frontend
        Web[Vite / Pure HTML]
        Desktop[Electron / Desktop App]
    end
    subgraph Backend Engine
        API[FastAPI Router]
        Worker[Neural Voice & Edge Workers]
    end
    subgraph Storage
        SQLite[(Local SQLite)]
        Sync[GitHub DB Synchronizer]
    end

    Web --> API
    Desktop --> API
    API --> Worker
    API --> SQLite
    SQLite --> Sync

1. Unified Backend Engine

LuxurAI runs a single backend server (main.py) powered by FastAPI that serves both the Web interface and handles local requests from the native Desktop App callback layers.

2. High-Accuracy AI Pipelines

Instead of relying on single LLM generations, our system delegates coding requests through a specialized multi-agent 3-AI chain (Gemini + DeepSeek). This chain automatically verifies and compiles patches to guarantee code correctness.

3. Neural Voice Engine (322 Voices / 142 Locales)

Voice synthesis workloads operate on a high-speed Edge-TTS and Cloudflare worker distribution layer. Generating human-like speech across 142 languages with sub-second latency, zero GPU overhead, and native sentence-boundary auto-chunking (type="chunk"). Audio is delivered directly to clients or played seamlessly through system speakers via the Python SDK.

4. Background DB Synchronization

Our local database state is continuously synced using a custom, secure GitHub storage integration. This acts as a robust backup and allows immediate restoration across redeployments on systems like Railway.