Projects

Leading research and development of a Random Forest machine learning model that automatically detects geomagnetic events with high accuracy—enabling early warning systems for space weather that protect satellite infrastructure worth billions of dollars. Currently improving model performance and working on publishing a comprehensive research paper that combines statistical validation with ML-based detection, advancing space weather forecasting capabilities.

• Developed and trained Random Forest classifier achieving 88.9% recall and 88.9% precision for SSC detection using engineered features from SYM/H magnetometer data—significantly outperforming traditional threshold-based methods.
• Engineered domain-specific features capturing pre-event quietness, compression amplitude, and ring current response to differentiate between storm commencements and impulses, enabling automated classification previously requiring manual expert analysis.
• Currently improving model performance for SSC/SI differentiation and multi-latitude generalization while preparing comprehensive research paper integrating machine learning detection with multi-latitude statistical validation.
• Implemented feature importance analysis revealing physical signatures—SYM/H minimum and pre-event quietness—as primary discriminators, validating model against known geophysical processes and demonstrating interpretable ML for space physics.

Leads product, data, and experience design for a platform that automates transaction insights, reveals anomalies, and empowers founders to trust their numbers. Enables startups to reconcile days of financial data in minutes, reducing manual accounting overhead and providing real-time visibility into spending patterns.

• Architected modular TypeScript services that ingest and classify financial events with explainable outcomes, processing thousands of transactions with high accuracy and providing actionable insights.
• Designed a polished interface that pairs luxury-inspired visuals with auditor-grade transparency, creating an intuitive experience that makes complex financial data accessible to non-technical founders.
• Built workflow automations enabling startups to reconcile days of data in minutes—reducing manual accounting time by over 90% and enabling faster financial decision-making.