Neighborhood Genie - HackUTD-2019
We addressed multiple issues in the housing market. Housing mortgages have become especially important in the home buying process. This is due to the increase in housing cost and relative stagnation in salaries across the board. A mix of Citi Bank, Fannie Mae, and JP Morgan's challenges, we deeply desired a unification of finance, and good for the world.
What it does
Our project implements both a home-buyer side and a mortgage-seller side.
The home-buyer is presented with a minimalist design questionaire in which they answer personality and preference based questions. Based on this input, the home-buyer is then matched with a neighborhood that best fits their results. The home-buyer can view this on a web-inbedded map, allowing for seamless, beautiful transition into their new community.
On the mortgage-seller side, in which the mortgage-seller can view an aggregated risk-assesment per user depending on location, and other risk-assessment factors. They get a data visualization for exquisite viewing purposes.
How we built it
Cameron Brill Lead back-end developer. Designed API and database, and redesigned file structure.
Dean Orenstein Lead front-end developer.
Ryan Hall Lead data analyst, built an OCR, and quality assurance.
William Deng Lead Python developer. Algorithmic processes and data visualizations.
Challenges we ran into
Getting APIs to behave.
OCR is a pain in the butt.
Communication was an issue.
Accomplishments that we're proud of
We are proud of making two coherent web apps that streamline neighborhood finding for potential home buyers and making mortgage risk assessment easier for financial institutions. We were also happy that we finally got things working. We had multiple ideas for the first few hours that never ended up taking off. We are proud that we decided on (two) an idea that is seamless and brilliant.
We are happy with our product and hope you all feel the same way.
What we learned
We also learned how to work more collaboratively overtime, which ultimately allowed us to avoid our demise from not finishing.
With more time, we would finish implementing live data, a more comprehensive web app, more beautiful data visualization with more data points and pivots, create a higher over-view of loans and mortgages (as we limited it to Dallas for testing purposes), and we would fully implement predictive modeling.