HEARTIROMER S. LOPEZ

Built and validated multi-timeframe EMA/StochRSI crypto trading strategies by migrating from an AC-CXT/pandas research pipeline to QuantConnect LEAN Python for production-grade backtesting and live execution, including checks for look-ahead bias and walk-forward leakage. Created a closed-loop daily trading workflow that reduced manual chart review time by about 25 hours per week using Claude-based morning market scans for TradingView chart analysis and subsequent post-trade validation, with audit logging to Google Drive. Diagnosed and fixed correctness issues in a multi-strategy backtesting system (e.g., drawdown aggregation, risk-manager bypass, PASS leakage, and profit-factor overstatement), improving out-of-sample profit factor by about 30%. Set up local AI agent tooling infrastructure (Jupyter MCP and GitHub MCP integrations) to support stable, repeatable development sessions with credential hygiene for secure tool-calling.

Delivered two production digital-banking platforms (SME Loan, FX Online) supporting working-capital financing and currency conversion. Led a cross-functional engineering team through the full Agile SDLC on AWS using C#, .NET Core, Angular 14, and Node.js, achieving an 80% reduction in loan-application processing time. Reduced delivery risk through structured risk identification and mitigation across sprint planning cycles, with zero critical production incidents post-launch. Improved stakeholder alignment and reduced rework via structured requirements gathering and status communication cadences, achieving a 30% reduction in requirement rework cycles.

Shipped enterprise REST API services for core banking functions, delivering 20 REST API endpoints and reducing response latency by 50%. Implemented Web APIs and Entity Framework Code-First data access in a C#/.NET Core/React stack. Cut deployment friction for containerized workloads by deploying to Azure Kubernetes Service (AKS) and contributing within Agile Scrum ceremonies, achieving a 60% faster release cycle.