🧠 TechScope AI

TechScope AI is an advanced AI-powered tech news dashboard that summarizes technology articles, scores their credibility, and extracts keywords. Built with FastAPI, SQLite, Streamlit, and HuggingFace Transformers, it provides a full-stack solution for tech news analysis.

---

Features

• ✅ Summarize tech articles using HuggingFace Transformers
• ✅ Credibility scoring for news articles
• ✅ Keyword extraction from article text
• ✅ FastAPI backend with API key authentication
• ✅ Streamlit frontend dashboard
• ✅ Persistent storage with SQLite (techscope.db)
• ✅ Fully RESTful API endpoints
• ✅ CORS-enabled for frontend integration
• ✅ Docker / Codespaces compatible

---

Tech Stack

• Backend: Python, FastAPI
• Frontend: Streamlit
• Database: SQLite
• NLP: HuggingFace Transformers, Sentence-BERT
• Other: Pydantic, Uvicorn, CORSMiddleware

---

Project Structure

TechScope-AI/
├── api/
│ ├── serve.py         # FastAPI backend
│ └── auth.py          # API key authentication
├── utils/
│ ├── summarizer.py    # Summarization utilities
│ ├── credibility.py   # Credibility scoring
│ ├── keywords.py      # Keyword extraction
│ └── save_data.py     # Load/save articles from DB
├── dashboard/
│ └── dashboard.py     # Streamlit frontend
├── data/
│ └── techscope.db     # SQLite database
├── templates/         # Optional HTML templates
├── static/            # Optional CSS/JS files
├── requirements.txt   # Python dependencies
└── README.md

---

Setup Instructions

1. Clone Repository

git clone https://github.com/Brahamanbtp/TechScope-AI.git
cd TechScope-AI

2. Create Virtual Environment

python3 -m venv venv
source venv/bin/activate  # Linux/macOS
# venv\Scripts\activate   # Windows

3. Install Dependencies

pip install -r requirements.txt

Dependencies include:

---

• fastapi==0.113.0
• uvicorn[standard]==0.23.2
• streamlit==1.28.0
• pydantic==2.7.1
• requests==2.32.0
• sentence-transformers==2.2.2
• transformers==5.7.0
• torch==2.2.0
• typing-extensions==5.3.0
• jinja2==3.1.3

---

4. Initialize SQLite Database

sqlite3 data/techscope.db

# Create summaries table if not exists:
CREATE TABLE IF NOT EXISTS summaries (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    summary TEXT,
    credibility REAL,
    keywords TEXT,
    created_at TEXT
);
Insert sample data (optional):

INSERT INTO summaries (summary, credibility, keywords, created_at)
VALUES
('OpenAI releases GPT-5 with enhanced reasoning capabilities and real-time memory.', 0.96, 'OpenAI,GPT-5,AI', '2025-07-22 10:00:00'),
('Google launches Gemini 2 to compete with ChatGPT in global markets.', 0.92, 'Google,Gemini,AI', '2025-07-21 14:30:00');

Running the Application

1. Start FastAPI Backend

python -m uvicorn api.serve:app --reload --host 0.0.0.0 --port 8000

---

Backend endpoints:

Method	Endpoint	Description
GET	/	Root, returns welcome message
GET	/articles	Load all stored articles
POST	/summarize	Summarize text (API key required)
POST	/credibility	Get credibility score (API key required)
POST	/keywords	Extract keywords (API key required)

---

2. Start Streamlit Dashboard

python -m streamlit run dashboard/dashboard.py  

Dashboard Features:

• Shows AI-summarized articles

• Displays credibility score and keywords

• Supports caching for faster updates

• User-friendly UI with Streamlit containers and expanders

API Key Authentication

• API key required for /summarize, /credibility, /keywords

• Verify with x-api-key header

• Implement your own API key in api/auth.py

Notes

• Backend uses CPU by default; can be changed to GPU if available

• FastAPI supports CORS, so frontend can be hosted separately

• Streamlit caching ensures minimal API calls

• SQLite database is lightweight and persistent for development

---

Troubleshooting

1. command not found errors for uvicorn/streamlit

python -m uvicorn api.serve:app --reload
python -m streamlit run dashboard/dashboard.py

2. No articles showing in dashboard

• Ensure techscope.db contains data in summaries table

• Ensure backend is running and API URL is correct in dashboard.py (API_URL)

3. Reinstall dependencies

pip install --upgrade -r requirements.txt

---

Future Enhancements

• Real-time news scraping and auto-summarization

• User authentication for personalized dashboards

• Multi-language support for articles

• Export articles and summaries to PDF/CSV

• Docker deployment for easy hosting

License

• MIT License

Author

Pranay Sharma