EspressoMind - Because research should be as stimulating as your morning coffee. ☕
"The scholar's AI research companion"

AI-Powered Research Assistant with Automated Source Verification

• Hybrid search across web + academic databases (arXiv, PubMed)
• Automatic source quality assessment with confidence scoring
• Dynamic research strategy selection based on query complexity

• APA-style citation generation
• Automatic reference section formatting
• Source metadata extraction (authors, dates, publishers)

• PDF text extraction with PyMuPDF
• Image OCR using Tesseract
• Web page scraping with Playwright

• Async pipeline for concurrent processing
• Intelligent caching of frequent queries
• Progressive result refinement

arxiv - Academic paper search
pymupdf - PDF processing
pytesseract - Image OCR
requests - HTTP client
python-multipart - File uploads

• Python 3.10+
• Playwright browsers (playwright install)
• Tesseract OCR (brew install tesseract on macOS)

# Clone repository
git clone https://github.com/dubeyakshat07/EspressoMind.git
cd EspressoMind

# Create virtual environment
conda create -n espresso python=3.10
# Install dependencies
pip install -r requirements.txt

uvicorn backend.main:app --reload --port 8000

Basic Text Query:

import requests

response = requests.post(
    "http://localhost:8000/analyze",
    json={
        "query": "Explain quantum entanglement",
        "depth": "balanced"  # quick/balanced/deep
    }
)

With PDF Upload:

with open('research.pdf', 'rb') as f:
    response = requests.post(
        "http://localhost:8000/analyze",
        files={
            'file': ('research.pdf', f),
            'file_type': (None, 'pdf')
        },
        data={
            'query': 'Summarize this paper'
        }
    )

{
  "answer": "Quantum entanglement is... [1][2]",
  "sources": [
    {
      "title": "Experimental observation of quantum entanglement",
      "url": "https://arxiv.org/abs/1234.5678",
      "source_type": "arxiv",
      "authors": ["Einstein, A.", "Podolsky, B."],
      "publish_date": "1935-05-15",
      "confidence": 0.92
    }
  ],
  "related_queries": [
    "Applications of quantum entanglement in computing",
    "Recent breakthroughs in entanglement research"
  ],
  "confidence_score": 0.87
}

EspressoMind/
├── backend/
│   ├── agents/
│   │   ├── research_agent.py  # Main research logic
│   │   └── tools.py          # Search/scraping tools
│   ├── schemas/
│   │   └── models.py         # Pydantic models
│   └── main.py               # FastAPI app
├── frontend/                 # Streamlit UI
│   └── app.py
├── tests/                    # Test suite
├── docs/                     # Documentation
├── requirements.txt          # Dependencies
└── README.md                 # This file

Edit .env file:

# Search Configuration
SEARXNG_URL=https://search.example.com
MAX_WEB_RESULTS=5
MAX_ACADEMIC_RESULTS=3

# AI Configuration
OLLAMA_MODEL=mistral
LLM_TEMPERATURE=0.3

# Performance
SCRAPE_TIMEOUT=20
PDF_PROCESS_TIMEOUT=30

Run the test suite:

pytest tests/ -v --cov=backend --cov-report=html

Key test coverage:

• API endpoints
• Search tools
• Document processing
• Citation generation

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Distributed under the MIT License. See LICENSE for more information.

Project Maintainer - [Akshat Dubey]