Developer: Divyansh Rai
University: IIT (BHU) VARANASI
Department: Chemical Engineering and Technology
"Revolutionizing personalized learning through autonomous AI agent collaboration"
Watch StudyBuddy AI in action! Access the complete demo video and project materials:
https://drive.google.com/drive/folders/1cjXftb8tCLZ-L_QKDgpx0V1FXDP0tn8U?usp=drive_link
What's included:
- 🎥 Complete demo video showing multi-agent system in action
- 🎯 Live plan creation and step execution demonstration
- ⚡ Bulk processing and real-time progress tracking
- 📊 Result display with study guides, flashcards, and quizzes
- 🔧 Technical implementation deep-dive
- 📋 Additional project documentation and materials
StudyBuddy AI is a sophisticated multi-agent artificial intelligence system that automates the entire learning workflow - from intelligent study plan generation to personalized content creation and progress tracking. Unlike traditional study tools, StudyBuddy employs three specialized AI agents that collaborate seamlessly to deliver a comprehensive, automated learning experience.
StudyBuddy implements a distributed AI agent system where each agent has specialized capabilities:
- 🎯 Planner Agent: Creates structured, goal-oriented study plans
- 🔬 Researcher Agent: Performs contextual research using RAG (Retrieval-Augmented Generation)
- ⚡ Executor Agent: Generates personalized learning materials and assessments
Problem: Traditional learning is inefficient - students waste hours planning what to study, researching topics manually, and creating study materials without clear structure or progress tracking.
Solution: StudyBuddy automates the entire learning pipeline through intelligent agent collaboration, reducing study preparation time by 80% while improving learning outcomes through personalized, structured content.
User Goal → Planner Agent → Structured Plan → Researcher Agent → Contextual Research → Executor Agent → Learning Materials → Results
- Backend: FastAPI (Python 3.8+) - High-performance async API
- Frontend: Streamlit - Interactive web interface with real-time updates
- Database: SQLite - Efficient data persistence and plan tracking
- AI Integration: Google Gemini 2.0 Flash - Advanced language model
- RAG System: Custom implementation with FAISS indexing
- Document Processing: ReportLab for PDF generation
- State Management: Session-based with progress tracking
# Intelligent study plan generation with tool assignment
class PlannerAgent:
def create_study_plan(self, goal: str) -> Dict[str, Any]:
# Analyzes learning goals and creates structured plans
# Assigns appropriate tools (RAG, Flashcards, Quiz, LLM)
# Generates unique step identifiers for tracking# Contextual research using RAG integration
class ResearcherAgent:
def research_step(self, step_description: str, step_tool: str) -> Dict[str, Any]:
# Generates targeted search queries
# Retrieves relevant content from knowledge base
# Provides contextual summaries for learning steps# Learning material generation and assessment creation
class ExecutorAgent:
def execute_step(self, step: Dict[str, Any], context: Dict[str, Any]) -> Dict[str, Any]:
# Creates study guides, flashcards, and quizzes
# Generates actionable learning artifacts
# Provides progress tracking and completion status- Intelligent Batching: Execute multiple study steps simultaneously
- Progress Tracking: Real-time visual progress indicators
- Error Handling: Robust failure recovery with detailed error reporting
- Result Aggregation: Comprehensive summaries of bulk execution results
- Responsive Design: Optimized for desktop and mobile devices
- Status-Based Theming: Color-coded visual indicators for step progress
- Interactive Components: Hover effects, smooth transitions, and intuitive controls
- Real-Time Updates: Live progress tracking and instant result display
- Professional PDF Generation: Publication-ready study plans with formatting
- Progress Preservation: Export includes completion status and results
- Shareable Content: Study materials optimized for collaboration
- Comprehensive Error Handling: Graceful failure management and user feedback
- API Documentation: Auto-generated OpenAPI/Swagger documentation
- Logging System: Detailed interaction logging for debugging and analytics
- Database Optimization: Efficient schema design with relationship management
- Python 3.8+ with pip
- Google Gemini API key (Get one here)
- Git for version control
# 1. Clone the repository
git clone https://github.com/divyansh-cyber/StudyBuddy.git
cd StudyBuddy
# 2. Create virtual environment (recommended)
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Configure environment variables
cp .env.example .env
# Edit .env file with your Google API key
# 5. Start the backend server
uvicorn backend.app:app --reload
# 6. Start the frontend (new terminal)
streamlit run frontend/streamlit_app.py
# 7. Access the application
# Navigate to http://localhost:8501 in your browserGOOGLE_API_KEY="your_gemini_api_key_here"
DATABASE_URL=sqlite:///./studybuddy.db
FAISS_INDEX_PATH=./faiss_index
DOCUMENTS_PATH=./documentsInput: "Learn machine learning fundamentals for data science projects"
Generated Plan:
1. 🔬 Research ML Concepts (RAG Tool)
→ Comprehensive overview of supervised/unsupervised learning
2. 📚 Create Concept Flashcards (Flashcards Tool)
→ 8 interactive flashcards covering key algorithms
3. ⚡ Practice with Examples (LLM Tool)
→ Hands-on coding exercises with real datasets
4. 📝 Assessment Quiz (Quiz Tool)
→ 7 questions testing ML understanding with explanations
Input: "Master React.js for frontend web development"
Generated Plan:
1. 🔬 JavaScript Prerequisites (RAG Tool)
→ ES6+ features, async programming, modern syntax
2. 📚 React Core Concepts (Flashcards Tool)
→ Components, JSX, hooks, state management
3. ⚡ Build Practice Projects (LLM Tool)
→ Todo app, weather dashboard, e-commerce features
4. 📝 React Ecosystem Quiz (Quiz Tool)
→ Testing knowledge of Router, Redux, testing frameworks
POST /api/plan
Content-Type: application/json
{
"goal": "Learning objective description"
}
Response: {
"plan_id": 123,
"goal": "Learning objective",
"plan": { /* Structured plan object */ },
"status": "created"
}POST /api/execute_step
Content-Type: application/json
{
"step_id": "plan_uuid_step_1"
}
Response: {
"step_id": "plan_uuid_step_1",
"status": "completed",
"result": { /* Learning materials */ },
"context": { /* Research context */ }
}POST /api/execute_steps_bulk
Content-Type: application/json
{
"step_ids": ["step_1", "step_2", "step_3"]
}
Response: {
"executed_steps": 2,
"failed_steps": 1,
"results": [ /* Array of results */ ],
"status": "completed"
}GET /api/download_plan_pdf/{plan_id}
Response: PDF file stream with proper headers# Run comprehensive feature tests
python test_new_features.py
# Run model validation tests
python test_models.py
# Debug system components
python debug_issues.py- ✅ API Endpoint Testing: All REST endpoints validated
- ✅ Multi-Agent Workflow: Agent collaboration tested
- ✅ Error Handling: Failure scenarios and recovery
- ✅ Performance Testing: Bulk execution and response times
- ✅ Integration Testing: End-to-end user workflows
- Response Time: Plan generation ~10-15 seconds
- Step Execution: Individual steps ~5-8 seconds
- Bulk Processing: 4-6 steps in ~30-45 seconds
- Success Rate: 95%+ completion rate for valid inputs
- Error Recovery: Graceful fallback mechanisms implemented
StudyBuddy/
├── 📁 backend/ # FastAPI server and agent logic
│ ├── app.py # Main FastAPI application
│ ├── planner.py # Planner Agent implementation
│ ├── researcher.py # Researcher Agent with RAG
│ ├── executor.py # Executor Agent for content generation
│ ├── llm.py # Google Gemini integration
│ ├── db.py # Database operations and ORM
│ └── 📁 tools/ # Specialized tools and utilities
│ ├── rag.py # RAG implementation with FAISS
│ └── calendar.py # Calendar integration utilities
├── 📁 frontend/ # Streamlit web interface
│ └── streamlit_app.py # Complete UI with modern styling
├── 📁 scripts/ # Utility and setup scripts
│ └── ingest_docs.py # Document ingestion for RAG
├── 📊 requirements.txt # Python dependencies
├── 📊System_design.txt # A documentation of the architecture, data design, component breakdown, chosen technologies, and reasons for them
├── 🧪 test_new_features.py # Comprehensive test suite
├── 🧪 test_models.py # Model validation tests
├── 🔧 debug_issues.py # Debugging and diagnostics
├── 🗄️ studybuddy.db # SQLite database (auto-generated)
└── 📄 README.md # This comprehensive documentation
- Innovation: First study application to implement specialized AI agent collaboration
- Technical Achievement: Designed distributed agent system with clear separation of concerns
- Impact: Enables more sophisticated learning workflows than single-AI approaches
- Innovation: Automatic selection of optimal learning tools based on content type
- Technical Achievement: Context-aware tool routing with fallback mechanisms
- Impact: Ensures appropriate learning modalities for different content types
- Innovation: Scalable batch execution with live progress tracking
- Technical Achievement: Asynchronous processing with user feedback integration
- Impact: Enables automation of complete learning sessions
- Innovation: Integration of retrieval-augmented generation for contextual learning
- Technical Achievement: Custom RAG implementation with document indexing
- Impact: Provides more accurate, relevant study materials
- Democratizes Quality Education: Makes structured learning accessible to everyone
- Reduces Learning Barriers: Eliminates manual planning and research overhead
- Supports Diverse Learning Styles: Multiple content formats (text, flashcards, quizzes)
- Time Optimization: Students focus on learning rather than preparation
- Consistent Quality: Ensures comprehensive topic coverage
- Progress Tracking: Clear visibility into learning advancement
- Educational Institutions: Integration with learning management systems
- Corporate Training: Professional skill development programs
- Personal Development: Self-directed learning initiatives
- Advanced RAG: Integration with larger document corpora
- Personalization: User learning style adaptation
- Multi-Modal Content: Support for images, videos, and interactive content
- Study Groups: Multi-user collaborative learning sessions
- Teacher Dashboard: Educator tools for curriculum management
- Progress Analytics: Advanced learning insights and recommendations
- LMS Integration: Canvas, Moodle, Blackboard connectivity
- Mobile Applications: Native iOS/Android apps
- API Ecosystem: Third-party developer tools and integrations
- Learning Insights: AI-powered learning pattern analysis
- Predictive Modeling: Success probability and intervention recommendations
- Performance Optimization: Continuous system improvement based on usage data
- FastAPI Expertise: Modern async Python web framework
- Database Design: Efficient schema design and ORM implementation
- API Architecture: RESTful design with comprehensive documentation
- Error Handling: Robust exception management and logging
- Large Language Models: Google Gemini integration and prompt engineering
- Multi-Agent Systems: Distributed AI architecture design
- RAG Implementation: Retrieval-augmented generation with vector databases
- Natural Language Processing: Text analysis and content generation
- Modern UI/UX: Responsive design with interactive components
- State Management: Complex application state handling
- Real-Time Updates: Live progress tracking and result display
- Cross-Platform Compatibility: Web-based solution with mobile optimization
- Scalable Architecture: Modular design supporting growth
- Performance Optimization: Efficient processing and caching strategies
- Security Considerations: API security and data protection
- Documentation: Comprehensive technical and user documentation
- Google Gemini AI: For providing advanced language model capabilities
- FastAPI Team: For the excellent async web framework
- Streamlit Team: For the intuitive UI development platform
- Open Source Community: For the foundational libraries and tools
StudyBuddy AI - Revolutionizing Education Through Intelligent Automation 🚀