System Architecture and Design

The Clover Server is designed to provide a robust and scalable backend for the Clover application. The architecture is based on a client-server model, where the server handles requests from clients, processes data, and communicates with external services as needed.

Components

Figure 1. Component Diagram of the Clover application.

Legend:
🖥️ UI 🧩 Editor 🌐 Backend 🤖 AI 🔐 Auth 🗄️ Database 📝 Logs

Backend

🌐 Gin (Golang)

The backend handles all API requests between the VS Code extension, AI model, and database.

🔹 Key Features:

• Processes Code Requests – Sends user code to the AI model and returns suggestions.
• Tracks User Behavior – Logs whether users accept, modify, or reject suggestions.

AI Model

🤖 Gemini (AI Code Generator)

The AI model generates code suggestions and sometimes introduces small mistakes to test users.

🔹 Key Features:

• Context-Aware Suggestions – Provides relevant recommendations based on the user’s code.
• Intelligent Mistakes – Occasionally tweaks suggestions with logical errors to test user focus.
• Adaptive Learning – Adjusts suggestions based on user behavior.

note

For more see AI Model.

Database & Authentication

🔐 Supabase (PostgreSQL, Auth)

Supabase handles user login and stores all interaction data.

🔹 Key Features:

• User Authentication – Manages logins and keeps track of individual progress.
• Logs User Activity – Records which suggestions were accepted, modified, or rejected.
• Real-Time Sync – Updates and injects user progress instantly to the AI model and on the dashboard.

note

For database schema design see Data Schemas.