Autonomous AI Trading System: A comprehensive Signal → Strategy → Execution pipeline integrated with a real-time React telemetry dashboard.
- Live Dashboard: View Deployment
- Backend API: View API Root
- Demo Video:
[INSERT YOUR YOUTUBE LINK HERE]
The system is built on a Modular Tri-Role Architecture ensuring separation of concerns and deterministic outcomes:
| Layer | Responsibility | Components |
|---|---|---|
| Role 3: Data & Signal | Market Ingestion | OHLCV Validator, Momentum & Sentiment Modules |
| Role 1: Strategy Architect | Decision Logic | Risk-Weighted Logic, Score Aggregator |
| Role 2: Execution Agent | Order Routing | 8-Point Safety Gate, Live State Tracking |
| Telemetry Dashboard | Visualization | Real-time PnL Tracking, Human-readable Trust Logs |
- Core: Python 3.11, FastAPI
- Data Science: NumPy, Pandas
- Protocol: JSON Structured Logging
- Framework: React.js (TailwindCSS)
- Visuals: Chart.js (Real-time Equity Curves)
- Networking: Axios
- Deployment: Railway (Full Stack)
- CI/CD: GitHub Actions
The Quant Telemetry Dashboard provides full transparency into the AI's "brain" in real-time:
- 📈 Live Market Execution: Dynamic visualization of Price Action vs. Portfolio Equity.
- 🧠 Trust Panel: A chronological audit trail explaining why the AI took a specific action.
- 📜 Trade History: Detailed logs including Signal type, Execution price, and individual trade PnL.
- 🛡️ Safety Status: Live monitoring of confidence scores and active risk parameters.
quant-architect-agent/
├── frontend/ # React.js SPA
│ ├── src/components/ # Dashboard, TrustPanel, ChartPanel
│ └── src/api/client.js # Centralized API Orchestration
├── backend/ # FastAPI Production Server
│ ├── strategy_math.py # Role 1: Deterministic Decision Logic
│ ├── execution/ # Role 2: Safety Gate & Order Routing
│ └── signals/ # Role 3: Technical & Sentiment Analysis
├── tests/ # 130+ Industrial-grade Unit Tests
└── run_demo.py # System Orchestrator
---
## 🛡️ Security & Safety Gates
Automated trading requires rigorous safety protocols. Our system includes:
* **Circuit Breakers:** The engine automatically halts after 3 consecutive losses to prevent algorithmic decay.
* **Strict Position Sizing:** Hard-capped at **25%** max allocation per trade.
* **Sanitized Logging:** Zero-trace logging ensures API keys and secrets never reach the console.
* **The 8-Check Gate:** Every trade must pass 8 validation points (Liquidity, Spread, Volatility, etc.) before execution.
---
## 🧪 Testing Suite
We maintain a high bar for reliability, ensuring every "Role" is battle-tested.
* **✅ Role 1 (Quant Engine):** 58/58 Tests Passed
* **✅ Role 2 (Execution):** 32/32 Tests Passed
* **✅ Role 3 (Signals):** 45/45 Tests Passed
```bash
# Run the full suite
pytest tests/
⚡ Quick Start
1. Installation
Bash
git clone [https://github.com/your-username/quant-architect-agent.git](https://github.com/your-username/quant-architect-agent.git)
cd quant-architect-agent
pip install -r requirements.txt
2. Environment Setup
Create a .env file in the root directory:
Code snippet
KRAKEN_API_KEY=your_key
KRAKEN_SECRET=your_secret
EXECUTION_MODE=dry_run
3. Execution
Bash
# Terminal 1: Backend
python main.py
# Terminal 2: Frontend
cd frontend
npm install && npm start
Submission Note for Lablab.ai: This project is a full-loop autonomous agent. Unlike basic bots, it validates signals against rigorous risk parameters, manages its own execution state, and provides a professional-grade telemetry dashboard for human oversight.