⚡ AlphaEngine - Algorithmic Trading Framework

Professional-grade Python framework for backtesting, optimizing, and deploying trading strategies.

---

🖥️ Dashboard Preview

Equity Curve & Metrics

Trade Visualization

Strategy Comparison

Parameter Optimization

Monte Carlo Stress Testing

---

What is AlphaEngine?

AlphaEngine is a complete algorithmic trading framework for serious quant traders. It includes 5 production-ready strategies, an interactive dashboard, walk-forward optimization, Monte Carlo simulation, and multi-exchange support.

This repo contains the free demo with 1 strategy (Momentum) and basic backtesting.

🔓 Free Demo (this repo)

• ✅ Momentum strategy
• ✅ 15 technical indicators (computed from scratch)
• ✅ Portfolio tracker with commission/slippage
• ✅ Performance metrics (Sharpe, Sortino, win rate, etc.)
• ✅ Position sizing
• ✅ Run provenance logging - every backtest captures the why, not just the what

🔒 Full Version (Get it here →)

• ✦ 5 strategies: Momentum, Mean Reversion, Breakout, RSI+MACD, Grid Trading
• ✦ Interactive Streamlit dashboard with 5 tabs
• ✦ Parameter optimization with Sharpe ratio heatmaps
• ✦ Monte Carlo simulation (1000+ scenarios, fan charts)
• ✦ Risk management suite: drawdown halts, trailing stops, Kelly criterion
• ✦ Multi-exchange: Binance (crypto) + Alpaca (US equities)
• ✦ MQL5 Expert Advisor (.mq5 file for MetaTrader 5)
• ✦ Telegram & Discord notifications

---

Quick Start

git clone https://github.com/Leotaby/alpha-engine.git
cd alpha-engine
pip install numpy pandas
python demo.py

Output:

╔══════════════════════════════════════════════════════════╗
║   ⚡  A L P H A  E N G I N E  —  Free Demo              ║
╚══════════════════════════════════════════════════════════╝

📊 Generated 1000 bars | $28,521 — $44,182

  ▸ Running Momentum strategy... ✓

════════════════════════════════════════════════════════════
  Total Return         +0.23%
  Sharpe Ratio         0.040
  Win Rate             100.0%
  Max Drawdown         2.05%
  Profit Factor        0.00
  Total Trades         2
════════════════════════════════════════════════════════════

  📝 Run saved → ID: f7e302ab  (logs/run_history.json)

────────────────────────────────────────────────────────
  Run ID       : f7e302ab
  Strategy     : momentum  |  BTC/USDT  |  1h
  Regime       : trending
────────────────────────────────────────────────────────
  HYPOTHESIS
  Default dual-EMA crossover with ADX > 25 to confirm trend
  strength. Baseline run before any parameter tuning.
────────────────────────────────────────────────────────
  PARAMETERS
    fast_period            12
    slow_period            26
    adx_threshold          25.0
────────────────────────────────────────────────────────
  RESULTS
    Total Return           +0.23%
    Sharpe Ratio           0.040
    Win Rate               100.0%
────────────────────────────────────────────────────────

  📋 Run History
  ID         Strategy    Regime      Return%  Sharpe  Trades
  f7e302ab   momentum    trending      +0.23   0.040       2

---

Run Provenance Logging

One of the hardest problems in systematic trading is remembering why you chose a specific configuration — not just what the parameters were. AlphaEngine solves this with a lightweight RunContext layer built into every backtest.

Every run captures:

ctx = RunContext(
    strategy_name="momentum",
    params={"fast_period": 12, "slow_period": 26, "adx_threshold": 25.0},
    hypothesis="ADX > 25 filters noise in ranging BTC markets. Testing baseline before tuning.",
    market_regime="trending",   # trending | ranging | volatile | mean-reverting
    symbol="BTC/USDT",
    tags=["baseline", "crypto", "adx-filter"],
    notes="Synthetic data — alternating trending/ranging regimes every 200 bars.",
)

Results are automatically persisted to logs/run_history.json. You can query your full history:

from core.run_logger import RunLogger

log = RunLogger("logs/run_history.json")

# Top 5 runs by Sharpe ratio
best = log.top_n(metric="sharpe_ratio", n=5)

# Filter by strategy and regime
trending = log.filter(strategy="momentum", regime="trending")

# Side-by-side parameter + result comparison
log.compare("f7e302ab", "a1b2c3d4")

The edge lives in the assumption that drove the setup, not just the parameters.

---

Add Your Own Strategy

Even in the free version, you can create custom strategies:

from strategies.base_strategy import BaseStrategy

class MyStrategy(BaseStrategy):
    name = "my_strategy"
    default_params = {"fast": 10, "slow": 30}
    
    @property
    def description(self):
        return "My custom crossover strategy."
    
    @property
    def param_grid(self):
        return {"fast": [5, 10, 15], "slow": [20, 30, 40]}
    
    def generate_signals(self, df):
        from utils.indicators import ema
        df = ema(df, self.params["fast"])
        df = ema(df, self.params["slow"])
        df["signal"] = 0
        df.loc[df[f"ema_{self.params['fast']}"] > df[f"ema_{self.params['slow']}"], "signal"] = 1
        df.loc[df[f"ema_{self.params['fast']}"] < df[f"ema_{self.params['slow']}"], "signal"] = -1
        return df

---

Full Version Pricing

	Starter	Pro ⭐	Premium
Price	$79	$129	$249
Strategies	2	5	5
Dashboard	Basic	Full (5 tabs)	Full
Optimizer	❌	✅	✅
Monte Carlo	❌	✅	✅
Risk Suite	Basic	Full	Full
Exchanges	1	2	2
MQL5 EA	❌	❌	✅
Support	Email	Priority	Video call

👉 Get AlphaEngine Pro →

---

Tech Stack

• Python 3.10+ - core framework
• Streamlit - interactive dashboard
• Plotly - professional charts
• NumPy/Pandas - data processing
• MQL5 - MetaTrader 5 Expert Advisor

---

Disclaimer

This software is for educational and research purposes only. It is not financial advice. Past performance does not guarantee future results. Always do your own research before trading with real money.

---

Built with 🧠 by a quantitative finance professional.