“The needs of the many” is a starting point. I build for the ones the system forgot to count.

I design tools for the communities most likely to be missed by conventional systems — Disabled people, low-income residents, chronically ill neighbors, and transit-dependent riders. Not as an afterthought. As the primary user.

Climate vulnerability follows the same maps. The neighborhoods most exposed to extreme heat, flooding, and environmental burden are the same ones conventional systems undercount. That's where I build.

By day I work in fintech. By conviction I work in civic tech.

Why I build: I nearly passed out walking my dog during Philadelphia’s June 2025 heat emergency. I’m Disabled. I don’t have the option of “just stay inside.” Neither do a lot of people.

My work lives in two tracks:

• Machine Learning for Equity: Supervised machine learning models built on public data, designed to help cities act before harm happens

• Vibe Coding for Climate Justice: No-cost, low barrier climate tools built with Claude assistance and open-source data, so communities don't need a development budget to get answers

I completed the Equitech Futures Civic Tech Institute (CTI) 2026 cohort. Think of me as an engineer who reads the mission statement first and the tech specs second.

I approach climate tech the same way I approach civic tech: equity isn't a filter you apply at the end. I work to center communities at the intersections of climate vulnerability and systemic neglect. Tools that don't address underserved resident groups don't really solve the problem.

No-cost, no-framework, no-barrier tools, built with AI assistance and open-source data. Every tool runs in the browser. No installs. No API keys. No budget required.

The goal: make public data legible and actionable for communities, advocates, and organizers who shouldn’t need a dev team to answer basic questions about their own neighborhoods. Each tool is built around one question a Philadelphian actually asks.

Every live tool is checked against WCAG 2.1 AA every week, and each one has a public accessibility statement listing known issues. See the latest results →

XGBoost · Supervised Classification · Climate Equity · Public Health

Assigns every US ZIP code a probability score for elevated heat illness during an approaching extreme heat event — so cities deploy cooling resources before hospitalizations pile up, not after.

The official US heat death count for 2023: 2,415. The estimated real count: ~11,000. That gap is a policy failure and a data problem. SustAInable is a proposal to close it.

• 🎯 Primary users: Municipal emergency managers, public health departments, community-based organizations
• 🗺️ Scope: National coverage at ZIP code level; deployable 48–72 hours before event peak
• 📊 Stack: XGBoost · CDC PLACES · NOAA HeatRisk · ACS 5-Year Estimates · SMOTE
• 🧪 Status: Design memo complete, exploratory analysis underway; seeking public health and emergency management pilot partners

XGBoost · Supervised Classification · Transit Equity · Disability Justice

Predicts which transit elevators, escalators, and platform lifts are likely to fail in the next 30 days — so maintenance happens before a Disabled rider shows up to a broken elevator with nowhere else to go.

Think of it as a credit score for mechanical equipment. One number per unit. One ranked list for maintenance teams. Zero riders stranded without warning.

• 🎯 Primary users: Disabled transit riders (and everyone else who can’t use stairs)
• 🗺️ Scope: Designed to scale from MTA NYC to any transit system on earth
• 📊 Stack: XGBoost · MTA Open Data · SHAP Explainability · Precision-Recall Optimization
• 🧪 Status: Full prototype pipeline built on synthetic data; seeking a transit agency partner to retrain on real maintenance records

• Transit Carbon Calculator expanded to 11 Pennsylvania transit agencies in partnership with Transit 4 All PA
• Accessible Voting Workshop (October 6, 2026), built around BallotGap PHL
• League of Women Voters of PA voter engagement event speaker (October 19, 2026)
• Equitech Futures Civic Tech Institute 2026 cohort, completed with two machine learning capstones

I don’t just build tools about civic systems. I work inside them.

I’m trying to make Philadelphia work better — for Disabled residents, transit riders, and communities that government data has historically undercounted.

Machine Learning XGBoost · Supervised Classification · SHAP Explainability
SMOTE · Imbalanced Class Handling · Precision-Recall Optimization

Civic Data CDC PLACES · NOAA HeatRisk · ACS 5-Year Estimates
MTA Open Data · OpenDataPhilly · CDC WONDER · NSSP

Vibe Coding Plain HTML/CSS/JS · Browser-based tools · AI-assisted development
Open-source data · Zero-dependency deployment

Domains Transit Equity · Disability Justice · Climate Equity
Public Health · Civic Technology · AI Governance

I am a lifelong comics reader, a Starfleet sympathizer, and someone who believes the Rebellion had better data infrastructure than the Empire ever gave them credit for. I watch Formula One for the telemetry. I’m Disabled and that shows up in how I build.

If you work in transit, public health, climate equity, or disability justice and you’re looking for someone who understands the policy and the pipeline — I want to hear from you.

Open to civic tech collaboration, transit agency pilots, fellowship opportunities, and conversations about building tools that center the people most affected by the systems they model.