Study Abroad AI Assistant - Project Documentation

Project Description

Ethiopian students often struggle to find accurate, up-to-date information about international study opportunities and scholarship programs. Study Abroad AI Assistant solves this by acting as an intelligent research assistant for international education.
It's a Retrieval-Augmented Generation (RAG) application that scrapes official scholarship websites and provides instant, LLM-powered answers with cited sources. This ensures accuracy and saves time, making it easier to explore programs, understand requirements, and plan study abroad journeys.

Core Functionality and Architecture

RAG Pipeline

Study Abroad AI Assistant implements the full RAG flow:

Document Ingestion

• Automated web scraping of Czech government scholarship websites
• Wikipedia integration for comprehensive country information
• Extracted data is stored with metadata (source, program type, country)

Embeddings & Retrieval

• Uses OpenAI text-embedding-ada-002 for semantic search
• Stores and retrieves documents via ChromaDB vector database
• Semantic similarity search across scholarship and country data

LLM Generation

• Answers are generated with Groq LLM for fast, high-quality responses
• Responses include source citations and confidence scores for traceability
• Chat history maintained for contextual conversations

User Experience (UX)

• FastAPI backend handles scraping, retrieval, and LLM pipeline
• RESTful API with session-based chat management
• Interactive API documentation via Swagger UI
• Persistent conversation history across sessions

Technical Stack & Setup

Languages & Frameworks: Python, FastAPI, SQLAlchemy, Alembic

Vector Store: ChromaDB

Embeddings: OpenAI text-embedding-ada-002

LLM: Groq

Database: PostgreSQL with asyncpg

Web Scraping: BeautifulSoup4, Requests

Deployment: Docker, Docker Compose

Setup (Local Development)

git clone https://github.com/kalebda/Uno.git
cd Assignment-Uno
cp env.example .env
# Edit .env and add your API keys:
# GROQ_API_KEY=your_groq_api_key_here
# OPENAI_API_KEY=your_openai_api_key_here

# Start all services
make dev

Environment variables:

GROQ_API_KEY - For LLM responses

OPENAI_API_KEY - For embeddings

POSTGRES_* - Database configuration (pre-configured)

API Usage Examples

Create New Chat Session

curl -X POST "http://localhost:8000/api/v1/new-chat" \
  -H "Content-Type: application/json" \
  -d '{
    "title": "Czech Scholarship Research",
    "country": "czech_republic"
  }'

Chat with AI Assistant

curl -X POST "http://localhost:8000/api/v1/chat" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "What are the requirements for Czech government scholarships?",
    "country": "czech_republic",
    "user_background": {
      "nationality": "Ethiopian",
      "education_level": "Bachelor"
    },
    "session_id": "your-session-id-here"
  }'

Access Points

• API Documentation: http://localhost:8000/docs
• Health Check: http://localhost:8000/health
• Database Admin: http://localhost:5050 (admin@admin.com / admin)

Data Sources

• Czech Government Scholarships: Official Ministry of Education website
• Country Information: Wikipedia integration for education systems, work opportunities, cities
• Programs Covered: Bachelor's, Master's, and PhD studies with requirements and deadlines

Key Features

✅ Session Management: Unique chat sessions with persistent history
✅ Source Attribution: All responses include clear source citations
✅ Confidence Scoring: Reliability indicators for response quality
✅ Multi-Program Support: Information about various scholarship programs
✅ Context Awareness: Follow-up questions maintain conversation context

Future Improvements

• Expand to additional countries beyond Czech Republic
• Add user authentication and personalized profiles
• Support file uploads for document processing
• Implement application tracking and progress monitoring
• Add mobile application interface

Summary

Study Abroad AI Assistant is a foundational RAG system that enables Ethiopian students to instantly query international scholarship information with source-backed responses. By combining web scraping, vector search, and LLM generation, it delivers practical AI assistance for international education planning with transparency and accuracy.
The system successfully demonstrates RAG implementation for real-world educational use cases, providing students with reliable, up-to-date information about study abroad opportunities while maintaining clear source attribution and ethical AI practices.