FYP BUDDY - AI-Powered Student Matching Platform

An intelligent platform that uses advanced AI agents to match university students for Final Year Projects (FYP) based on their skills, interests, project ideas, and compatibility metrics. Built with modern technologies including LangGraph agents, React, and FastAPI.

🎯 Project Overview

FYP BUDDY solves a common problem in universities: students struggling to find compatible partners for their Final Year Projects. Our AI-powered system analyzes student profiles, project ideas, skills, and interests to suggest optimal team formations.

Core Problem Solved

Manual Partner Search: Students waste time manually searching for project partners
Skill Mismatches: Teams form without considering complementary skills
Interest Alignment: Students with different project interests end up in same groups
Project Feasibility: Teams lack clarity on project scope and technical requirements

Our Solution

AI-Powered Matching: LLM-based compatibility scoring across 5+ criteria
Comprehensive Profiling: Skills, interests, project ideas, and academic background
Smart Recommendations: Ranked list of most compatible potential partners
Project Generation: AI-generated project ideas tailored to student backgrounds

🚀 Key Features

For Students

Intuitive Profile Creation: Multi-step form with real-time validation
Smart Matching Algorithm: AI analyzes compatibility across multiple dimensions
Project Inspiration: AI-generated project ideas matching your interests
Partner Discovery: Find students with complementary skills and shared interests
Contact Information: Direct access to matched students' contact details

For Institutions

Automated Team Formation: Reduce administrative overhead
Better Project Outcomes: Higher success rates through better team matching
Analytics & Insights: Track matching effectiveness and student preferences
Scalable Solution: Handle large cohorts efficiently

🏗️ System Architecture

┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   React Frontend │────│   FastAPI Backend │────│   MongoDB Atlas │
│   (User Interface) │   │   (API Gateway)   │   │   (Data Storage) │
└─────────────────┘    └──────────────────┘    └─────────────────┘
                                │
                                ├── LangGraph Agents
                                │   ├── Project Generation
                                │   ├── Interest Profiling  
                                │   └── Match Finding
                                │
                                ├── Redis Queue
                                │   └── Background Jobs
                                │
                                └── External APIs
                                    ├── Groq LLM
                                    └── LangSmith Tracing

Technology Stack

Frontend (React)

React 19.1.1 with modern hooks
Vite for lightning-fast development
Tailwind CSS for responsive design
Real-time form validation

Backend (FastAPI)

Python 3.12+ with FastAPI framework
LangGraph for AI agent orchestration
MongoDB for scalable document storage
Redis for background job processing

AI/ML Stack

LangChain for LLM integration
Groq for high-performance inference
LangSmith for observability
Custom multi-criteria matching algorithm

Infrastructure

Docker containers with multi-stage builds
Azure Container Apps for scalable deployment
MongoDB Atlas for managed database
Redis for job queuing

🧠 AI Matching Algorithm

Our sophisticated matching system uses specialized LangGraph agents:

Core Matching Agent (Connection Finding)

The heart of the platform - analyzes student compatibility using a 5-Criteria Scoring System (0-5 scale):

Idea Similarity: How well project ideas align
Interest Match: Compatibility of student interests
Shared Interests: Common areas of focus
Skill Complementarity: How skills complement each other
Overall Compatibility: Holistic assessment

Advanced Features:

Handles missing data gracefully
Considers interdisciplinary projects
Ranks results by compatibility score
Provides explainable matching decisions

Data Generation Agents (Development & Testing)

For populating the system with realistic test data:

Project Generation Agent

Creates diverse, feasible FYP project ideas
Generates realistic student profiles with projects
Ensures department-skill alignment
Used for system testing and demonstration

Interest Profiling Agent

Generates varied student interest profiles
Creates realistic skill-interest combinations
Populates database with diverse student backgrounds
Supports comprehensive matching algorithm testing

📊 Data Models & Validation

Student Profile Structure

{
  "id": "22K-1234",
  "personal_info": {
    "department": "Computer Science",
    "batch": 2022,
    "cgpa": 3.5,
    "email": "[email protected]"
  },
  "project_details": {
    "title": "AI Healthcare System",
    "domain": "Healthcare AI",
    "idea": "Develop an AI system for medical diagnosis...",
    "tech_stack": ["Python", "TensorFlow", "React"]
  },
  "compatibility_data": {
    "skills": ["Machine Learning", "Web Development"],
    "interests": ["AI", "Healthcare", "Data Science"]
  }
}

Validation Rules

Student ID: University format (22K-1234)
Email: Must use @nu.edu.pk domain
CGPA: Range 2.0-4.0 with realistic distribution
Skills/Interests: Minimum 1 each, maximum flexibility
Project Scope: Feasible for 3-person undergraduate team

🔄 User Journey

1. Profile Creation (3-5 minutes)

Registration → Personal Info → Skills & Interests → Project Ideas → Review → Submit

2. AI Processing (2-10 minutes)

Profile Ingestion → Interest Analysis → Compatibility Scoring → Match Ranking

3. Results & Connection (Instant)

View Matches → Contact Details → Team Formation → Project Kickoff

📁 Project Structure

fyp-buddy/
├── README.md                    # This file - Project overview
├── frontend/                    # React frontend application
│   ├── README.md               # Frontend-specific documentation
│   ├── src/components/         # React components
│   ├── package.json           # Frontend dependencies
│   └── vite.config.js         # Build configuration
├── app/                        # FastAPI backend application
│   ├── README.md              # Backend-specific documentation
│   ├── main.py                # FastAPI application entry
│   ├── src/agent/             # AI agents and core logic
│   ├── test_scripts/          # Comprehensive test suite
│   ├── Dockerfile             # Container configuration
│   └── pyproject.toml         # Backend dependencies
└── docs/                      # Additional documentation (if any)

🚀 Quick Start

Prerequisites

Node.js 18+ and Python 3.12+
MongoDB (local or Atlas)
Redis (local or cloud)
API Keys: Groq and LangSmith accounts

1. Clone Repository

git clone <repository-url>
cd fyp-buddy

2. Backend Setup

cd app/
cp .env.example .env          # Configure API keys
pip install -r pyproject.toml # Install dependencies
uvicorn main:app --reload     # Start backend

3. Frontend Setup

cd frontend/
npm install                   # Install dependencies
npm run dev                   # Start frontend

4. Access Application

Frontend: http://localhost:5173
Backend API: http://localhost:8000
API Docs: http://localhost:8000/docs

📈 Performance & Scale

Current Capabilities

Throughput: 1000+ student profiles
Match Processing: 2-10 minutes per query
Concurrent Users: Scalable with Redis queuing
Accuracy: ~85% satisfaction rate in initial testing

Scaling Strategy

Horizontal Scaling: Stateless backend design
Caching: Redis for frequent queries
Database: MongoDB sharding for large datasets
CDN: Asset delivery optimization

🧪 Testing & Quality

Comprehensive Test Suite

Located in app/test_scripts/ with 8 comprehensive tests:

Configuration validation
Database connectivity
AI agent functionality
Error handling scenarios
End-to-end workflows

Quality Metrics

Test Coverage: 90%+ for critical paths
Performance Tests: Response time validation
Integration Tests: Full workflow validation
Error Handling: Graceful failure modes

🌟 Use Cases & Applications

Academic Institutions

Computer Science Departments: FYP team formation
Engineering Schools: Project partner matching
Bootcamps: Cohort project assignments
Research Labs: Collaboration facilitation

Beyond Academia

Hackathons: Team formation based on skills
Corporate Projects: Cross-functional team building
Open Source: Contributor matching
Startup Incubators: Co-founder matching

🔮 Future Enhancements

Planned Features

Connecting groups with supervisors: teams can find their supervisors
Advanced Analytics: Team success prediction
Integration APIs: LMS and university system integration
Mobile Application: Native iOS/Android apps

AI Improvements

Personality Matching: Myers-Briggs compatibility
Learning Style Analysis: Complementary learning approaches
Success Prediction: Historical team performance analysis
Agent to work on project ideas: Coming up and refining project ideas
Agent to act as another member: Will be the 4th member of the project

🤝 Contributing

We welcome contributions from the community! Here's how to get started:

Development Setup

Follow the Quick Start guide above
Read component-specific READMEs:
- Frontend Documentation
- Backend Documentation

Contribution Guidelines

Fork the repository
Create a feature branch
Test thoroughly using the test suite
Document your changes
Submit a pull request

Areas for Contribution

UI/UX improvements
AI algorithm enhancements
Performance optimizations
Test coverage expansion
Documentation improvements

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

👨‍💻 Developer

Developed by: Bilal Asif Burney
Email: [email protected]
GitHub: [GitHub Profile]

🔗 Component Documentation

For detailed setup, API documentation, and component-specific information:

🎨 Frontend Documentation: frontend/README.md
- React setup and configuration
- Component architecture
- User interface details
- Deployment instructions
⚙️ Backend Documentation: app/README.md
- FastAPI setup and configuration
- AI agent architecture
- Database schema and operations
- API endpoint documentation
- Testing procedures

Ready to revolutionize student team formation? Start with the component READMEs above, or jump straight into the Quick Start guide!

Built with ✨ for the love of creation and innovation

Name		Name	Last commit message	Last commit date
Latest commit History 39 Commits
.github/workflows		.github/workflows
app		app
frontend		frontend
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md

License

BilalAsifB/fyp_buddy

Folders and files

Latest commit

History

Repository files navigation