DongshinBuddy: AI-Powered Campus Guide 🎓🤖

🌟 Overview

DongshinBuddy is an intelligent campus companion application built specifically for the students and faculty of Dongshin University. By integrating cutting-edge AI models (Gemini & Groq) and real-time cloud synchronization, the app simplifies campus navigation, provides instant academic assistance, and centralizes university resources into a single, sleek interface.

📥 Download Now

🖼️ App Screenshots: A Quick Tour

Take a look at the core functionalities and interface of DongshinBuddy:

Home & AI Chat

🚀 Features

🧠 Smart AI Assistant

• Dual-Engine Intelligence: Powered by Google Gemini and Groq Cloud (LPU) for lightning-fast responses.

• Campus-Aware: Pre-configured with university-specific context to answer academic and administrative queries accurately.

• Markdown Support: Rich text rendering for better readability of AI-generated tables, lists, and code snippets.

📍 Interactive Campus Map

• Building Locator: Custom pins for every major building on the Dongshin University campus.

• Seamless Navigation: One-tap integration with Kakao Map via Intent API for precise walking or driving directions.

• Lightweight WebView: Optimized map rendering for smooth performance on all devices.

🔗 Quick University Portal

• Centralized Links: Instant access to the Student Portal, LMS, Library, and Course Schedules.

• Real-time Updates: Link data is managed via Firebase, allowing for instant updates without requiring a new app release.

🎨 Modern UI/UX

• Material Design 3: Follows the latest Google design guidelines for a premium feel.

• Dynamic Theme: Full support for Dark Mode and System Default themes.

• Privacy-First: No login required, no data collection, and 100% anonymous AI interactions.

---

🛠 Technology Stack

• Language: Kotlin (Native Android)

• Architecture: MVVM (Model-View-ViewModel)

• AI Engines: Google Gemini API & Groq Cloud (LPU)

• Backend: Firebase Realtime Database

• Networking: Retrofit, OkHttp, Coroutines & Flow

• UI/UX: Material Design 3 (M3), Lottie Animations

• Rich Text: Markwon (For AI Markdown rendering)

• Navigation: Kakao Map Intent API & WebView

🏗 System Architecture

The app follows a clean MVVM architecture to ensure the codebase is scalable and maintainable.

1. View: Handles UI rendering (Activities/Fragments).

2. ViewModel: Manages UI states and acts as a bridge to data.

3. Repository: Orchestrates data flow between AI APIs and Firebase.

4. Remote Data: Fetches intelligence from Groq/Gemini and metadata from Firebase.

🧠 RAG (Retrieval-Augmented Generation)

DongshinBuddy utilizes Dynamic RAG by fetching university-specific context from Firebase Realtime Database to ensure the AI provides up-to-date campus information.

• Knowledge Source: Verified campus data (schedules, building details, portal links) is stored and managed in Firebase.

• Dynamic Retrieval: The AppRepository fetches the latest context from Firebase during the app session.

• Context Injection: The ChatViewModel injects this real-time data into the AI prompt via AIConfig.

• Benefit: This allows for "Over-the-Air" (OTA) updates to the AI's knowledge base without requiring a new app update on the Play Store.

📂 Project Structure

com.bayazidht.dongshinbuddy
├── api             # Retrofit Client and Groq/Gemini Service interfaces
├── data
│   ├── local       # Local datasets (Chips, Context, and Links)
│   └── repository  # Logic for fetching data (AIConfig, AppRepository)
├── model           # Data classes (ChatMessage and API response models)
├── ui
│   ├── activities  # Main and Chat screens
│   └── fragments   # Home and Settings fragments
├── adapter         # RecyclerView adapter for the Chat interface
├── viewmodel       # ViewModel for managing UI state and AI logic
└── utils           # Helper classes (Prefs, Constants, and Custom Tabs)

---

📄 Project Resources & Documentation

For a deeper dive into the technical details and quality assurance of DongshinBuddy, please refer to the documents below:

• Software Requirement Specification (SRS) – Detailed functional and non-functional requirements.

• Software Design Specification (SDS) – Technical architecture, database schema, and API designs.

• Testing & Quality Report – Summary of closed testing (12+ testers) and AI response verification.

---

📦 Installation & Setup

1. Clone the Repository:

   Bash

   git clone https://github.com/bayazidht/DongshinBuddy.git
   
   

2. API Keys: Create a secrets.properties or add your keys to BuildConfig:

   • GROQ_API_KEY=your_key_here

   • GEMINI_API_KEY=your_key_here

3. Firebase Setup: Add your google-services.json to the app/ directory.

4. Build: Sync with Gradle and run on an Android device (API 24+).

---

👨‍💻 Development Team

• Syed Bayazid Hossain – Developer & System Architect

• Md Rizvee Hasan – Data Research & Analytics

• Published By: Coder Spark

---

📄 License & Credits

• Copyright: © 2026 Coder Spark. All rights reserved.
• License: Distributed under the MIT License.
• Acknowledgments: Special thanks to Dongshin University for the campus data and the developers of Groq and Google Gemini for providing the AI infrastructure for this project.