The Challenge

I wanted to build something in AI that wasn't just a demo notebook — software that could look at a real image, identify what's in it, and show that result in a way anyone could read. The hard part wasn't only getting detection working. It was making it presentable, and getting comfortable putting it in front of people.

What I Built

Object detection software: a build focused on getting detection functional, then refining the aesthetics of how results are displayed so the output reads clearly instead of looking like raw model output. I iterated on the visuals alongside the logic, because a tool nobody can interpret isn't finished.

My Role

Solo build — model and detection logic, interface aesthetics, and the presentation of the work itself.

Skills Used

Computer Vision · Object Detection · AI Development · Machine Learning Concepts · Visual Design · Technical Presenting

What's Next

Making it bigger. Dennine's session pushed me to think about reach — how a sidequest becomes something brands, companies and wider audiences actually use. That's the direction I'm taking it.

The Challenge

BKLVLUP is building Brooklyn's first Community Resilience Hub on Utica Avenue. Most resilience hub surveys are English-only, written by professionals for professionals, and never reach the residents they claim to represent. Before construction, before programming, the community needed to be heard.

My Role & Contribution

I built and deployed the Haitian Creole translation of the Community Resilience Hub survey — tested for reading level and format, not just translated. I also designed the concept for a 'Heat & Air Risk Check' tool: residents enter their block and get a plain-language summary of heat vulnerability, air quality and flood risk, with practical steps they can actually take.

The Data That Changed the Design

Heatstroke is the leading environmental cause of death for New York City seniors. That single finding moved cooling and elder access from nice-to-have to must-design-for — and shaped how simple I made the interface.

Impact

The survey is live in English, Spanish and Haitian Creole and collecting responses. Feedback was gathered in person at events including National Night Out at the Shirley Chisholm Center, and the research now feeds the Hub's design priorities as it moves from concept to construction.