AI Trading Platform with Strategy Backtesting & Market Analysis

Project Overview We are looking for an experienced AI/ML and Quantitative Developer to build an AI-powered quantitative trading and market-analysis platform for the Indian stock market. The goal is to create a scalable system that combines market data, quantitative analysis, algorithmic trading strategies, historical backtesting, machine learning, news/sentiment analysis, and AI-powered market intelligence. This will initially be developed as an MVP, with the architecture designed so it can later be expanded into a full SaaS platform. Core Requirements 1. Market Data Integration - Integrate market-data APIs such as DhanHQ or equivalent. - Collect and process historical and, where available, live market data. - Support stocks and major Indian indices. - Support multiple timeframes, including intraday data. - Build a reliable data ingestion and storage pipeline. 2. Quantitative Analysis & Feature Engineering - Develop technical and mathematical market features. - Momentum, trend, volatility, volume, price-action and statistical features. - Create a modular feature-engineering framework that allows new features to be added easily. 3. Trading Strategy Framework - Develop a modular framework for implementing and testing trading strategies. - Initially support a large collection of strategies across: - Trend following - Momentum - Mean reversion - Breakout - Volatility - Statistical strategies - Strategies should have configurable parameters and a standardized interface. 4. Backtesting Engine The platform should be able to automatically backtest strategies on historical data and generate metrics such as: - Total Return - CAGR - Sharpe Ratio - Sortino Ratio - Maximum Drawdown - Win Rate - Profit Factor - Number of Trades - Average Trade Return - Risk/Reward metrics The system should support parameter optimization and proper train/test or out-of-sample validation to help reduce overfitting. 5. Machine Learning / AI Strategy Selection Develop an ML-based component that can analyze market conditions and historical strategy performance to identify or rank potentially suitable strategies. Possible inputs include: - Market regime - Volatility - Trend strength - Volume - Technical features - Historical strategy performance The objective is to create an intelligent strategy-ranking/selection layer rather than relying on one fixed trading strategy. 6. News & Sentiment Analysis - Collect relevant financial news. - Perform NLP-based sentiment analysis. - Generate sentiment scores for stocks/indices. - Combine sentiment information with quantitative market analysis. 7. Fundamental Analysis Integrate a reliable financial-data source to provide company fundamentals such as: - Revenue - Earnings - EPS - P/E - P/B - ROE - Debt - Growth metrics - Other relevant financial ratios 8. AI Market-Intelligence Chatbot Build an AI chatbot that can answer questions related to: - Stocks and indices - Market conditions - Quantitative analysis - Fundamentals - News and sentiment - Strategy performance - Backtesting results RAG/vector search can be used where appropriate. Technology Preferred technologies include: - Python - Pandas / NumPy - Scikit-learn - PyTorch/TensorFlow where required - FastAPI - PostgreSQL/TimescaleDB or equivalent - Redis where useful - Vector database - LLM/RAG frameworks - React/Next.js or equivalent for the frontend - AWS/cloud deployment We are open to recommendations from the developer if a different technology provides a better architecture. Ideal Freelancer You should have strong experience in several of the following: - Quantitative finance - Algorithmic trading - Backtesting - Python - Machine learning - Time-series analysis - Financial APIs - Data engineering - NLP/LLMs - RAG systems - Production AI applications Experience with Indian stock-market APIs such as DhanHQ, Zerodha, Angel One, etc. is a plus. Deliverables - Market-data ingestion pipeline - Quantitative feature-engineering system - Modular trading-strategy framework - Backtesting engine - ML-based strategy-selection component - News/sentiment analysis - Fundamental-analysis module - AI/RAG market chatbot - Basic web dashboard/API - Database architecture - Automated testing - Documentation - Deployment-ready code Important This is intended to become a long-term product, not just a one-off research notebook. Clean architecture, reproducibility, testing, documentation, and scalability are important. Please include examples of quantitative trading, backtesting, AI/ML, or financial-data projects you have previously worked on. When applying, briefly explain how you would approach the architecture and which components you would build first. Please do not apply if you have only built basic trading indicators or simple Python backtesting scripts. We are looking for someone capable of developing a complete AI/quantitative system.