Build Algo using Python for Intraday Trading Strategy
Budget: $250 – $750 USD
I already run this multi-layer intraday trend-and-momentum strategy live on TradingView, and it behaves exactly as intended there. Now I need the same logic translated to a Python algo that talks to Zerodha’s Kite API so it can place and manage real NSE/BSE orders automatically.
All of the trading logic is ready in Pine Script: dual-timeframe directional scoring, twin SuperTrend confirmation, jump/flip pattern triggers, momentum state scoring, ATR-managed exits, dynamic stop-losses, a trend filter, and strict intraday time gating. The job here is to reproduce that behaviour faithfully in Python, keep it modular for easy parameter tweaks, and wire it for live execution.
I want you to build on familiar tools—Pandas for data handling, TA-Lib for indicator parity, and Backtrader for back-testing. During development I will supply the original Pine code and walkthroughs of each module so you can mirror calculations line by line.
Deliverables I expect
• A well-documented Python package (or notebook set) that connects to Zerodha, streams intraday data, executes the strategy, and handles order/state persistence
• Backtrader test harness showing identical entry/exit points to the TradingView reference
• Built-in generation of trade logs (CSV or SQLite is fine), performance metrics (CAGR, hit rate, max-DD, etc.), and optional Matplotlib/Plotly visual charts of equity curve and parameter heatmaps
• A quick-start readme plus comments that let me adjust position sizing, time-gates, and risk parameters without touching core logic
Acceptance criteria: on at least two weeks of historical 1-minute data, the Python back-test must match the original Pine results within normal rounding tolerance; in paper trading the script should submit, modify, and exit orders on Zerodha without manual intervention.
If you have proven Indian-market algo builds on Zerodha, message me with a brief outline of similar projects and the average live latency you achieved. Only seasoned algo coders, please.
All of the trading logic is ready in Pine Script: dual-timeframe directional scoring, twin SuperTrend confirmation, jump/flip pattern triggers, momentum state scoring, ATR-managed exits, dynamic stop-losses, a trend filter, and strict intraday time gating. The job here is to reproduce that behaviour faithfully in Python, keep it modular for easy parameter tweaks, and wire it for live execution.
I want you to build on familiar tools—Pandas for data handling, TA-Lib for indicator parity, and Backtrader for back-testing. During development I will supply the original Pine code and walkthroughs of each module so you can mirror calculations line by line.
Deliverables I expect
• A well-documented Python package (or notebook set) that connects to Zerodha, streams intraday data, executes the strategy, and handles order/state persistence
• Backtrader test harness showing identical entry/exit points to the TradingView reference
• Built-in generation of trade logs (CSV or SQLite is fine), performance metrics (CAGR, hit rate, max-DD, etc.), and optional Matplotlib/Plotly visual charts of equity curve and parameter heatmaps
• A quick-start readme plus comments that let me adjust position sizing, time-gates, and risk parameters without touching core logic
Acceptance criteria: on at least two weeks of historical 1-minute data, the Python back-test must match the original Pine results within normal rounding tolerance; in paper trading the script should submit, modify, and exit orders on Zerodha without manual intervention.
If you have proven Indian-market algo builds on Zerodha, message me with a brief outline of similar projects and the average live latency you achieved. Only seasoned algo coders, please.
Related categories:
JavaScript
Python
Software Architecture
Financial Analysis
SQLite
Data Analysis
Trading
API Integration
Backtesting
Pandas