README.md

metadata

title: ECM Quant AI
emoji: 🚀
colorFrom: gray
colorTo: yellow
sdk: docker
pinned: false
app_port: 7860

ECM Quant AI | Analyst Dashboard

ECM Quant AI is a professional-grade quantitative pricing engine. Originally prototyped in Streamlit, it has been re-architected as a high-performance FastAPI web application to meet production latency requirements.

It features a "Goldman Sachs" style analyst dashboard using server-side rendering (Jinja2) and lightweight vanilla JavaScript for interactive charting.

---

🚀 Key Features

• FastAPI Backend: High-performance asynchronous endpoints for market data processing.
• Production Dashboard: Custom HTML/CSS/JS frontend (no heavyweight frameworks) for maximum speed and "Human-Written" quality.
• Real-Time Signals: Calculates Momentum, Volatility, and Beta against the S&P 500 (^GSPC) using yfinance.
• Institutional Aesthetic: Dark mode with Gold (#FFD700) accents.
• Zero-Keys: Fully operational using public market data rails.

🛠️ Usage

Local Development

1. Install dependencies:

   pip install -r requirements.txt
   

2. Run the server:

   uvicorn main:app --reload
   

3. Access Dashboard: Open http://127.0.0.1:8000 in your browser.

Docker Deployment

The project is containerized for Hugging Face Spaces (Docker SDK).

docker build -t ecm-quant-ai .
docker run -p 7860:7860 ecm-quant-ai

📊 Methodology

The engine normalizes 6-month historical price data to derive pricing recommendations:

1. Momentum (30d): Rolling rate-of-change vs Benchmark.
2. Volatility: Annualized standard deviation.
3. Pricing Recommendation: Heuristic model f(momentum, volatility) -> [Low, High] range.

📂 Project Structure

├── main.py                 # FastAPI Application (Entry Point)
├── templates/
│   └── index.html         # Jinja2 Dashboard Template
├── static/
│   ├── style.css          # CSS Variables & Theme
│   └── script.js          # Client-side Charting (Plotly)
├── requirements.txt       # Dependencies
├── Dockerfile             # Uvicorn container
└── README.md              # Documentation

---

Built for the Modern ECM Desk.