Nathaniel Mareano

• Designed and built a dashboard-planning application adopted by the Solutions team; connected discovery requirements, demonstration datasets, metric definitions, calculations, visualization plans, and reusable guidance in one workspace.
• Built a Python MCP server with nine tools for dataset profiling, formula validation, source-based claim verification, and dashboard guidance.
• Created executive analytics demonstrations using contact-center data, speech topics, AI prompts, sentiment, and quality scoring to answer customer-service and business questions.
• Developed KPI calculations to investigate data-quality issues and operational inefficiencies, identify revenue opportunities, and recommend changes. Contributed to first-contact-resolution analysis involving repeat contacts, escalations, quality scores, KPI tiles, and drilldowns.
• Used Claude Code and Codex to support implementation, review, and testing while owning the application’s architecture, analytical logic, validation, and iteration with team feedback.
• Prepared code, datasets, dashboard plans, metric build sheets, setup instructions, and handoff materials so the Solutions team could understand and continue the work.

• Built Excel cost models comparing on-premises application and server costs with Azure deployment scenarios to support infrastructure planning and IT budgeting.
• Mapped applications and their dependencies in Ardoq and organized project data in SharePoint to support enterprise-architecture work.
• Supported early-stage AI proofs of concept across IT teams.

• Researched government IT markets and analyzed requests for information and proposals (RFIs and RFPs) to assess bid and teaming opportunities.
• Used Python and Excel to automate research and internal data analysis.

• Translated statements of work and performance work statements (SOWs and PWSs) into planning summaries.
• Built Python automation prototypes and documented technical workflows in internal guides.

Sports research application

• Developing an iOS and web product to make sports research freely accessible, using SwiftUI, React, Python, FastAPI, and SQLite.
• Built data workflows covering NFL, NBA, and MMA information, organizing team and athlete statistics with source attribution. The current working web prototype focuses on NFL research, with visible calculations, sample limits, and source freshness.

City-data and simulation project

• Developing a city-scale data and simulation prototype for Harrisonburg, connecting public geography, traffic-volume records, and aggregate population inputs.
• Built an interactive map, source registry, and intersection inspector to make observed data and modeled movement inspectable. The next stage is to validate network behavior and connect vehicle movement to signal logic before comparing traffic strategies.

Project contributions

• Built CSM, an offline Python prototype for medical-invoice reference review, with local data management, searchable results, and evidence connecting procedure and diagnosis pairs to applicable CMS Article versions.
• Packaged and tested a native desktop workflow with review history and exports using synthetic invoice data. The application supports reference screening and human review.
• Contributed data-integration and AI-assisted analysis approaches to a federal transportation-data RFI response.