Cut insurance-claim denials by 36% by rebuilding the invoice-submission validation pipeline and hardening error handling, reducing support tickets. Built data-driven “Nudge” logic scoring GAD-7/PHQ-9 and historical results to notify therapists at the right time. Fixed GraphQL cache invalidation with targeted clear rules to keep responses fresh. Secured form submissions with encryption and a controlled decryption workflow for the data team. Shipped real-time dashboards for invoice volume, validation errors, and latency with anomaly alerts, shrinking detection from hours to minutes.

Ran a full-stack codebase audit that uncovered security gaps and performance bottlenecks; delivered a remediation roadmap that cut response times by up to 75% and eliminated all high-severity vulnerabilities. Designed robust database schemas and RESTful endpoints, employing dependency injection for maintainability and testing. Orchestrated feature delivery that supported a ~2000% surge in adoption by RIA firms.

Collaborated on internal C#/TypeScript applications for major Rockstar titles, enhancing player progression and economy features to boost engagement. Redesigned the player-engagement app to leverage real-time analytics, improving the timing and relevance of marketing campaigns. Built a shared TypeScript/React component library for consistency and velocity. Consolidated SQL databases to unify data for marketing and cross-platform analytics, and authored clear technical docs (Markdown/Mermaid) to streamline rollouts.

Modernized browsing and checkout by refactoring legacy code and introducing React, Next.js, and GraphQL. Managed a team of three to overhaul global UI components and built a bespoke CMS for streamlined content management. Led performance work—caching, image optimization, lazy loading, CSS refinements—resulting in faster page loads and higher user satisfaction.

Founded an ETF comparison site that helps investors analyze holdings, fees, and fundamentals side-by-side. Built with Next.js/React/TypeScript and Node.js, containerized with Docker, and deployed on AWS. Implemented server-rendered pages, caching, and a MySQL-backed data pipeline to ingest and normalize ETF datasets. The site serves ~3,000 monthly visitors.

Node.js script that screens and auto-allocates Prosper loans using ~200 data points with configurable filters and position sizing. Supports fully automated reinvestment and hands-off operation. Personal results have averaged ~7.2% vs a platform average ~5.3% (past performance not indicative of future results).