- DIY and Home Improvement
- Backpacking and Hiking
- Historical Exploration
- Adventure Sports

Data Scientist and AI Engineer with over a decade of expertise spanning traditional ML, analytics, and modern agentic AI systems. Deep specialization in financial services—from credit decisioning and model risk management to cutting-edge LLM applications. Proven track record of bridging classical machine learning rigor with emerging AI technologies while driving cross-functional teams to deliver high-impact solutions in complex, regulated environments.
• Agentic AI Systems: Architected and deployed an AI Underwriting Agent for consumer finance decisioning, defining evaluation-first design (4-stage testing framework covering hallucination prevention, risk-flag precision, data grounding), integrated Langfuse/HoneHive observability, and authored Model Risk Assessment documentation with quantified impact analysis and phased rollout strategy aligned to governance requirements.
• Production ML & Model Development: Designed and deployed geo-based risk binning model integrating multiple scoring outputs (application score, early default, metascore) to optimize origination strategies at SA4 granularity.
• Model Governance & Monitoring: Led quarterly oversight and performance monitoring of personal lending, credit card, customer and collections models. Implemented daily score integrity checks, directed calibration investigations to resolve performance mismatches, and enhanced monitoring frameworks for early degradation detection and regulatory compliance.
• Cross-Functional Strategy & Risk Management: Orchestrated scorecard enhancement initiatives, data pipeline development, and Model Risk Assessments across multiple portfolios. Coordinated data discovery and access alignment across product, engineering, risk, and operations teams, and partnered with engineers on observability platform integration via PR reviews and acceptance testing.
Credit Risk Modeling: Developed customer scorecard for Citibank (USA) using logistic regression, leading data exploration, feature engineering, and statistical validation (significance testing, correlation analysis, stability metrics). Enhanced model interpretability and documentation for regulatory compliance and stakeholder alignment.
• Data Engineering & Automation: Re-engineered SAS macros with embedded QA frameworks to improve data pipeline reliability and reduce manual rework. Established automated quality checks that became standard practice, accelerating model development cycles and improving data integrity.
• Healthcare Analytics Transformation: Migrated manual MS Access/SAP reporting workflows for Presbyterian Health Services (USA) to automated SAS EG pipelines. Built interactive Tableau dashboards combining descriptive and predictive analytics, reducing report generation time from days to hours.
• Self-Service Analytics: Engineered web application using SAS stored procedures enabling business users to dynamically query and analyze reporting outputs, democratizing data access