Quantitative Trading Research Platform

Master Director: Most is already built: Project $10-$30 MAX FIRM PROJECT: Autonomous Non-Print Quantitative Research Platform NOTE: Most of this framework is already built. I am seeking an experienced Python/QuantConnect LEAN developer to complete, integrate, enhance, and automate the remaining components. Please read the attached project overview before bidding. OBJECTIVE Build a modular quantitative research platform that transforms Interactive Brokers (IBKR) Time & Sales data into statistically validated trading strategies. This is NOT a traditional trading bot. The objective is to create a repeatable research pipeline that can continuously generate, validate, rank, deploy, and improve trading strategies from proprietary Non-Print market structure data. The entire system must run locally on Windows using Python and be designed with low RAM and CPU usage. WORKFLOW The completed workflow should operate automatically: 1. Download historical IBKR Time & Sales data. 2. Build Bid and Ask Non-Print Structure Engines. 3. Store all structural events in a replayable database. 4. Generate 100 independent Multi-Resolution Line Break Structure Engines. 5. Calculate quantitative research features. 6. Export standardized research datasets. 7. Automatically process those datasets through my existing QuantConnect LEAN Bridge. 8. Generate candidate strategies. 9. Validate and rank every strategy. 10. Deploy only approved strategies. 11. Monitor live performance. 12. Retrain research whenever strategy performance degrades. The system should be fully modular so every stage can be improved independently. PROJECT MODULES 1. OpenClaw AI (Master Controller) Acts as the master research manager. Responsibilities include: • Control the entire workflow • Launch each module • Verify successful completion • Maintain configuration • Track historical strategies • Compare new vs previous strategies • Monitor live performance • Detect degradation • Trigger retraining • Enforce prop firm rules • Generate reports and logs OpenClaw manages the research process rather than generating trading strategies itself. 2. Non-Print Data Engine Using IBKR Time & Sales (ib_insync): Capture: • Timestamp • Last Price • Trade Size • Bid Price / Size • Ask Price / Size Create two independent engines: • Bid Non-Print Engine • Ask Non-Print Engine Detect and store: • Non-Print Events • Liquidity Voids • Skipped Prices • Structural Gaps • Void Persistence • Structural Velocity • Structural Acceleration Use incremental processing only. Historical data should be stored in PostgreSQL or TimescaleDB. 3. Market Structure Engine Automatically generate: 100 independent Line Break engines Resolution: 1 Line Break / 1 Tick through 100 Line Break / 1 Tick Each engine maintains independent structural state. The objective is allowing research to determine which structural resolution currently has the highest statistical edge. 4. Feature Engineering Automatically calculate research features including: • Structural Velocity • Structural Acceleration • Void Size • Void Persistence • Reversal Frequency • Persistence Length • Compression • Expansion • Resolution Agreement • Resolution Divergence • Structural Volatility • Trend Persistence • Liquidity Imbalance • Directional Consensus All feature calculations should be documented and exported. 5. QuantConnect LEAN Research Automatically consume exported datasets. Current LEAN Bridge already performs: • Feature Engineering • Strategy Generation • Parameter Optimization • Walk Forward Validation • Monte Carlo Validation • Parameter Stability • Strategy Ranking The search should be guided by my trading framework rather than completely random exploration. 6. Validation & Risk Automatically validate: • Walk Forward • Monte Carlo • Out-of-Sample • Parameter Stability • Drawdown • Sharpe • Sortino • Profit Factor • Win Rate • Expectancy • Recovery Factor Reject statistically weak strategies automatically. 7. Live Strategy Manager Deploy approved strategies. Continuously monitor: • Drawdown • Win Rate • Expectancy • Performance degradation Automatically disable deteriorating strategies and activate stronger validated strategies. 8. Prop Firm Risk Manager Continuously enforce account rules before orders are submitted. Monitor: • Daily Profit Goal • Maximum Drawdown • End-of-Day Drawdown • Position Size • Holding Time • Position Scaling • Rule Compliance GENERAL REQUIREMENTS Python Async architecture Event-driven processing Replayable database CSV export REST API WebSocket API Structured logging Low RAM usage Low CPU usage Incremental processing Every module should remain independently testable and replaceable. Please review the attached detailed project specification before bidding. I am looking for someone with strong experience in Python, quantitative research, Interactive Brokers API, event-driven systems, databases, and preferably QuantConnect LEAN.

Project ID: 40569256