• Hi, I'm Justin

    I'm a Software Engineer, Data Scientist, and User Researcher, focusing on transportation/mobility, diasability, education, and civic-oriented applications.

    In my free time, I love ameateur photography, poetry/spoken word, new music, spikeball, and exploring parks & public transit.

    LinkedIn

    My work experience includes data design & engineering in big tech, CS education design at Stanford, large-scale outreach and research for the US Senate, and ML for anti-bias and local business development in Oakland & Senegal. Currently serving on the Board of Directors @ Develop for Good

Recent Projects

Optimizing Bus Routes for equity using Reinforcement Learning

Transportation Policy, Data Science

Detecting Bias in Oakland's Parking Enforcement

Transportation Policy, Data Science

Predicting Bus Delays using Graph Neural Networks

Transportation, Data Science

NLP-Assisted research tool for local journalists

Civic Data Science, User Research

Designing for ADHD: UI/UX for a time management app

User Research, Disability Design

Optimizing Bus Routes for equity using Reinforcement Learning

In 2023, Yoo & Lee proposed an alternate approach to bus route route design that optimizes for the connectivity of bus-dependent riders.
This brief paper provides a reinforcement-learning based implementation and evaluation on a selection of major U.S. cities using census and GTFS.

Detecting Bias in Oakland's Parking Enforcement

A study completed in August 2022 for the City of Oakland’s Department of Transportation (OakDOT). In response to citizen concerns about unequal allocation of parking enforcement resources, I created a combined dataset to evaluate the influence of demographic factors on citations issued in 2019.
Over 100 metrics, including racial makeup, income, education, 311 reports, road paving quality, home ownership, parking availability, and vehicle ownership were analyzed to determine their effect on enforcement.

Predicting Bus Delays using Graph Neural Networks

Medium write-up and code samples for a Graph Neural Network model to predict bus delay times.

An NLP-Assisted research tool for local journalists

User research, data engineering, interface design, and final implementation for a tool to provide local journalists with NLP-assisted search, entity extraction, and topic-tracking against legislation, meeting minutes, and agendas from local municipalities. Final project for Stanford CS206: Computational Journalism.

Designing an app for ADHD time management

Problem space exploration, user interviews, design evolution, and prototyping for a time management app, optimized for adults with ADHD. Final project for Stanford CS147: Human Computer Interaction Design.