🚀 A next-generation, self-evolving AI trading system for Quotex with advanced transformer models, emotional intelligence, and autonomous decision-making capabilities
NEXUS represents the convergence of cutting-edge AI, financial engineering, and autonomous systems. Built from the ground up as a fully dynamic, self-evolving trading intelligence, NEXUS adapts to market conditions in real-time using advanced transformer architectures, emotional state modeling, and sophisticated exploration-exploitation strategies.
- 🧠 Self-Evolving Intelligence: Real-time learning and adaptation using transformer models and reinforcement learning
- 🎭 Emotional Awareness: Advanced emotional state tracking (greed, fear, confidence) that influences trading decisions
- 🔬 Exploration-Exploitation Balance: Sophisticated epsilon-greedy strategies with confidence and uncertainty metrics
- 🛡️ Safety-First Design: Multiple safety guards, payout thresholds, and risk management layers
- 🎨 Modular Architecture: Every component is replaceable, extensible, and independently testable
- 🔐 Security & Privacy: All data remains local by default with encrypted configuration support
graph TB
subgraph "🧠 Intelligence Layer"
T[Transformer Models]
RL[RL Agents]
RD[Regime Detection]
E[Exploration Controller]
end
subgraph "⚙️ Core Engine"
NE[NexusEngine]
ES[Emotional State]
RM[Risk Management]
MS[Meta Strategy]
end
subgraph "🔌 Adapters"
QA[Quotex Adapter]
PA[Payout Manager]
CA[Catalog Manager]
end
subgraph "🖥️ Interface"
GUI[PySide6 GUI]
CLI[CLI Interface]
end
T --> NE
RL --> NE
RD --> NE
E --> NE
NE --> ES
NE --> RM
NE --> MS
NE --> QA
QA --> PA
QA --> CA
GUI --> NE
CLI --> NE
- 🔥 Transformer Models: State-of-the-art attention mechanisms for market prediction
- 🎯 Meta-Strategy Coordination: Intelligent ensemble of multiple AI models
- 🧪 Reinforcement Learning: Adaptive agents that learn from market feedback
- 📊 Regime Detection: Automatic identification of market conditions
- 🎲 Exploration Controller: Dynamic epsilon computation with confidence metrics
- 💰 Payout Guards: Configurable thresholds with override capabilities
- 📈 Emotional Intelligence: Greed, fear, and confidence tracking
- ⚖️ Advanced Risk Sizing: Emotion-based position sizing with equity limits
- 🚨 Panic Controls: Emergency stop mechanisms and trade limits
- 📝 Audit Trails: Complete logging of all decisions and overrides
- 🖥️ Rich GUI Dashboard: Real-time monitoring with PySide6
- ⚡ CLI Interface: Powerful command-line tools for automation
- 📊 Live Statistics: Performance metrics, emotional state, and exploration rates
- 🎛️ Dynamic Controls: Real-time parameter adjustment without restarts
- 📦 UV Package Management: Lightning-fast dependency resolution
- 🧪 Comprehensive Testing: 22 passing tests covering core functionality
- 📚 Rich Documentation: Detailed specs, architecture docs, and API references
- 🔍 Type Safety: Full mypy compatibility with strict typing
- 🚀 Hot Reloading: Dynamic configuration updates
- Python 3.13.6+ (required for advanced type hints and performance)
- UV Package Manager (recommended for fastest setup)
- Windows/Linux/macOS (cross-platform support)
# Clone the repository
git clone https://github.com/ItzSwapnil/nexus.git
cd nexus
# Create virtual environment with UV (recommended)
uv venv .venv
source .venv/bin/activate # Linux/macOS
# OR
.venv/Scripts/Activate.ps1 # Windows PowerShell
# Install dependencies
uv pip install -e .
# Verify installation
pytest -q # Should show "22 passed"# 🖥️ Launch GUI Dashboard
nexus
# OR
python -m nexus.main
# ⚡ CLI Interface
nexus-cli
# 🧪 Development Mode
python main.py| Component | Status | Description |
|---|---|---|
| 🧠 Core Engine | ✅ Stable | Complete engine with emotional state & registries |
| 🎯 Payout Guards | ✅ Production Ready | Threshold enforcement with audit overrides |
| 🎭 Emotional Intelligence | ✅ Active | Greed/fear/confidence tracking & decision influence |
| 🔍 Exploration Controller | ✅ Implemented | Epsilon computation with confidence metrics |
| 🖥️ GUI Dashboard | ✅ Functional | Real-time monitoring & controls |
| 📊 Market Catalog | ✅ Simulated | In-memory market data with OTC flags |
| 🧪 Testing Suite | ✅ Complete | 22 comprehensive tests |
| 🤖 Transformer Models | 🚧 In Development | Advanced architecture scaffolded |
| 🎮 RL Agents | 🚧 Prototype | Basic framework implemented |
| 🔌 Live Trading | Safety-first development approach |
NEXUS uses a sophisticated configuration system with environment variable support:
# config.yaml
quotex:
email: "your-email@example.com"
password: "your-password"
demo_mode: true
trading:
payout_threshold: 80.0
base_trade_amount: 5.0
max_risk_per_trade_percent: 2.0
default_asset: "EURUSD"
auto_trade_enabled: false
exploration:
base_epsilon: 0.15
k_uncertainty: 0.35
min_epsilon: 0.01
max_epsilon: 0.6
ai:
enable_gpu: true
batch_size: 256
learning_rate: 0.001# Override any config with environment variables
export NEXUS__TRADING__PAYOUT_THRESHOLD=85.0
export NEXUS__AI__ENABLE_GPU=false# Run full test suite
pytest -v
# Run specific test categories
pytest tests/test_engine.py -v # Core engine tests
pytest tests/test_intelligence.py -v # AI component tests
pytest tests/test_gui_smoke.py -v # GUI smoke tests
# Performance testing
pytest tests/test_e2e_playwright.py -v # End-to-end testsCurrent Test Coverage: 22 passing tests covering:
- ✅ Core engine functionality
- ✅ Emotional state management
- ✅ Risk management systems
- ✅ Payout guard mechanisms
- ✅ Configuration validation
- ✅ GUI component initialization
- ✅ Intelligence module integration
sequenceDiagram
participant GUI as 🖥️ GUI Dashboard
participant Engine as ⚙️ NexusEngine
participant AI as 🧠 AI Models
participant Guard as 🛡️ Payout Guard
participant Quotex as 🔌 Quotex API
GUI->>Engine: Trade Signal Request
Engine->>AI: Get Market Prediction
AI->>Engine: Confidence + Direction
Engine->>Engine: Update Emotional State
Engine->>Engine: Calculate Position Size
Engine->>Guard: Check Payout Threshold
alt Payout OK
Guard->>Engine: Trade Approved
Engine->>Quotex: Execute Trade
Quotex->>Engine: Trade Result
Engine->>Engine: Update Statistics
Engine->>GUI: Display Results
else Payout Low
Guard->>Engine: Trade Blocked
Engine->>GUI: Show Warning
end
- Multi-head attention for pattern recognition
- Positional encoding for temporal awareness
- Adaptive learning rates and regularization
- Real-time inference optimization
# Emotional state influences trading decisions
emotion_state = {
"greed": 0.7, # Increases position sizes
"fear": 0.3, # Reduces risk exposure
"confidence": 0.8 # Affects exploration rate
}Dynamic epsilon computation balancing:
- Confidence metrics: Sharpe ratio, stability, win rate
- Uncertainty factors: ATR, model disagreement, spreads
- Payout optimization: Higher payouts reduce exploration
The system tracks comprehensive performance indicators:
| Metric | Description | Current Target |
|---|---|---|
| 📊 Sharpe Ratio | Risk-adjusted returns | > 1.5 |
| 📉 Max Drawdown | Largest peak-to-trough decline | < 5% |
| 🎯 Win Rate | Percentage of profitable trades | > 60% |
| ⚡ Execution Speed | Average trade execution time | < 100ms |
| 🧠 Model Accuracy | Prediction accuracy rate | > 70% |
nexus/
├── 🧠 intelligence/ # AI models & algorithms
├── ⚙️ core/ # Engine & execution logic
├── 🔌 adapters/ # External API integrations
├── 🖥️ gui/ # PySide6 interface
├── 🛡️ payouts/ # Safety & validation
├── 📊 catalog/ # Market data management
├── 🔧 utils/ # Configuration & logging
└── 🧪 tests/ # Comprehensive test suite
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Run tests (
pytest -v) - Commit changes (
git commit -m 'Add amazing feature') - Push to branch (
git push origin feature/amazing-feature) - Open a Pull Request
⚠️ SIMULATION ONLY: NEXUS v2.0.0 is currently in simulation mode. No real money trading is active.
🛡️ Risk Warning: Trading involves substantial risk. Past performance does not guarantee future results.
🔒 Security: Keep your credentials secure. NEXUS stores sensitive data locally only.
- 📖 Architecture Guide - System design & components
- 🎮 GUI Manual - Interface usage & features
- ⚙️ API Reference - Technical specifications
- 🚀 Usage Guide - Trading strategies & tips
License: MIT License - see LICENSE file for details
Support:
- 🐛 Bug Reports: GitHub Issues
- 💬 Discussions: GitHub Discussions
🌟 Star this project if you find it useful! 🌟
Built with ❤️ by the NEXUS Team
