Urban Planner & Data Analyst
I analyze US cities using Python and open data — uncovering inequities in city services, tracking disaster risk, mapping economic patterns, and benchmarking digital governance — to help cities become smarter, more equitable, and more resilient.
I hold a BA and MA in Urban Planning and am completing my Master of Science in Urban Informatics at Northeastern University. My work sits at the intersection of city policy, geospatial analysis, and data engineering.
I built this portfolio to demonstrate how Python and open data can answer real urban questions — from which neighborhoods wait longest for city services, to which states face the greatest climate disaster burden.
My goal is to become a smart city architect who helps municipalities make evidence-based decisions that improve quality of life for all residents.
Smart Cities
Analyzes 100,000+ Boston 311 service requests to reveal which neighborhoods face the longest response times — and what that means for urban equity and smart city investment priorities.
Public Safety
Examines Boston Police, EMS, and Fire Department incident data to map hotspots, track COVID's impact on emergency call patterns, and identify resource allocation gaps across neighborhoods.
Climate Analytics
Tracks national air quality trends across all 50 states using EPA AQI data, identifying where air quality is improving and where climate-driven wildfire smoke is reversing decades of clean-air progress.
Mobility
Analyzes MBTA and national transit ridership trends from 2018 to 2023 to quantify COVID's impact on public transportation and model the pace of ridership recovery across US metro areas.
Economic Analysis
Uses Census County Business Patterns data across all 50 states to map business density, quantify small business dominance, and reveal the urban-rural economic divide in productivity and payroll per establishment.
Disaster Risk
Analyzes 70,000+ FEMA disaster declarations from 1953 to 2024 to identify which states carry the greatest disaster burden, how disaster frequency is accelerating, and what COVID-19 meant for emergency management at scale.
Open Data
Scores US federal and state governments on open data transparency — measuring dataset quantity, format diversity, topic coverage, and accessibility. USGS publishes 3× more datasets than any other federal agency.
Climate Justice
Analyzes heat island intensity across 20 US cities using regression, correlation heatmaps, and ridgeline distributions to show how green space, income, and density interact with urban temperatures.
Predictive Modeling
Builds a 7-predictor OLS regression model on 240 metro-area observations to quantify how transit, walkability, green space, commute, and schools drive home values across Tier 1 and Tier 2 metros.
Mobility Equity
Examines Bluebikes trip patterns with a heatmap of 3.9M rides, exposes station access inequities across income quintiles, and measures how the e-bike program democratized mobility in underserved neighborhoods.
Health Equity
Tests the park-health hypothesis across 30 US states using Pearson correlation, quartile violin plots, and bubble charts to show how park access, income, and chronic disease burden intersect.
Sustainability
Tracks America's energy transition with stacked area charts, slope charts, and a composite efficiency index that scores 25 states on per-capita consumption, renewable share, and smart grid investment.
Streamlit App
A deployable Streamlit web application that turns 60,000 Boston 311 service records into an interactive equity dashboard. Four tabs — neighborhood rankings, an interactive Mapbox map, income-tier trend lines (2019–2024), and a service heatmap — all driven by sidebar filters that update every chart instantly. The data generator encodes a real equity gradient: response times scale with neighborhood income, producing the 2× gap that mirrors the actual Boston data.
SQL / Database
Builds a production-grade SQLite database from 60,000 Boston 311 requests across
three normalized tables, then writes 10 analytical SQL queries that reveal the
equity story hidden in the data. Each query introduces a new technique —
correlated subqueries for the equity gap, RANK() OVER (PARTITION BY income_tier)
for within-group rankings, LAG() for year-over-year change, and a rolling
3-month average to surface seasonal rodent activity spikes.
Machine Learning
Asks a research question — do US cities form natural groupings beyond geography? — then answers it with an unsupervised ML pipeline: StandardScaler normalizes 8 urban metrics across 50 cities, K-Means (k=4, validated by elbow + silhouette analysis) finds the clusters, and PCA reduces them to 2D for visualization. The result: four distinct archetypes — Dense Transit Hubs, Coastal Tech Cities, Emerging Sun Belt, and Industrial Midwest — each with a distinct policy profile.
REST API
A production-ready REST API exposing urban metrics for 18 US cities — the same backend powering the City Search widget on this page. Built with FastAPI for async performance and automatic OpenAPI documentation. Endpoints cover transit scores, walkability, green space, energy profiles, housing prices, and a server-side equity index computed on every request. CORS-enabled so any frontend can call it; structured so a PostgreSQL swap-in requires minimal changes.
Spatial GIS
Applies real GIS operations — not just mapping — to the question of park equity
in Boston. Neighborhoods are projected from WGS84 to UTM Zone 19N (EPSG:32619)
so that a buffer(500) means exactly 500 meters. A unary_union()
merges all park catchment zones, then intersection().area calculates the
precise percentage of each neighborhood within walking distance of green space.
The Pearson r = 0.51 correlation with income suggests proximity alone doesn't
explain the gap — access barriers matter too.
Search any US city to pull live data from my deployed FastAPI backend — transit scores, walkability, green space, and a real-time equity index calculated server-side.
GET /cities/{city}
Powered by FastAPI · Deployed on Railway
Two comprehensive, beginner-friendly tutorials — one for all 7 Python projects and one for all 5 R/RStudio projects. Covering environment setup, pandas, ggplot2, statistical methods, SQL, machine learning, REST APIs, GIS, and policy interpretation. Free on GitHub.
I'm actively looking for opportunities in urban tech, smart city consulting, civic data, and urban informatics research. If you're working on making cities smarter and more equitable, I'd love to connect.