An academic project showcase platform powered by Django — now enhanced with Google Gemini via Vertex AI for smarter project exploration.

VidyaVault is a Django-based platform that allows students to upload and showcase their final-year or academic projects. It enables users to explore peer submissions, receive structured faculty evaluations, and view performance through dashboards and leaderboards.

💡 In this GenAI-enhanced version, VidyaVault integrates a Gemini-powered chatbot (via Vertex AI) that helps users understand each project by analyzing its README and code. It acts as a learning companion for students, giving instant explanations and summaries.

• 📤 Upload and manage academic/final-year projects
• 🔍 Explore peer submissions with filters and search
• 🧑‍🏫 Faculty scoring system
• 📈 Dashboard for participation and tech usage
• 🏆 Leaderboard based on scores and likes
• 💬 Comment system with threaded replies
• 🤖 Gemini AI assistant: Ask questions about the project

VidyaVault includes an AI assistant on each project page using Google’s Gemini API:

This integration enhances the learning experience by enabling contextual Q&A without altering the original project code.

• Framework: Django 4.2
• Language: Python 3.10+
• Database: SQLite (dev) / PostgreSQL (prod)
• Auth: Django's User model + custom roles
• AI API: Google Gemini via google-generativeai Python SDK
• .env: Securely stores GOOGLE_API_KEY

• Bootstrap 5 for layout and responsive UI
• Font Awesome for icons
• Django Templates for page structure
• JavaScript: Chatbot UI and fetch logic
• Markdown support using marked.js in chatbot replies

• Charts: Department-wise participation
• Tech stack usage visualization
• Skills word cloud from project tags

1. User clicks "Ask AI About This Project"
2. Enters a question related to the project (e.g., "What does this project do?")
3. Django backend builds a prompt using:
   • Static instruction
   • README/code content (retrieved or embedded)
4. Gemini responds with an explanation or breakdown
5. UI shows formatted response with Markdown support

vidyavault/
├── manage.py                   # Django project starter
├── .env                        # Contains secrets like GOOGLE_API_KEY
│
├── vidyavault/                 # Project settings and URL configurations
│   ├── __init__.py
│   ├── settings.py
│   ├── urls.py
│   └── wsgi.py
│
├── projects/                   # Main Django app
│   ├── models.py               # Project, Score, Comment models
│   ├── views.py                # Views for listing, detail, upload, etc.
│   ├── chatbot_gemini.py       # Gemini API integration
│   ├── urls.py                 # App-specific URLs
│
├── templates/
│   ├── base.html               # Base layout
│   ├── registration/          # Auth pages
│   │   ├── login.html
│   │   └── register.html
│   └── projects/              # Project-related templates
│       ├── home.html
│       ├── about.html
│       ├── dashboard.html
│       ├── leaderboard.html
│       ├── profile.html
│       ├── project_detail.html
│       ├── project_list.html
│       ├── upload_project.html
│       ├── admin_review.html
│       ├── score_project.html
│
└── media/                      # Uploaded project files

# Clone repo
git clone https://github.com/NIJOY-P-JOSE/VidyaVault.git
cd VidyaVault

# Install dependencies
pip install -r requirements.txt

# Create a .env file with your Google Gemini API key
echo "GOOGLE_API_KEY=your_api_key" > .env

# Run server
python manage.py runserver

• Name: Nijoy P Jose
• Email: [email protected]