Designed user-friendly, interactive charts to help track bookings, revenue, and utilization of events

Built REST API using Express and Node.js, stored event data with Redux, and implemented frontend designs in React and Chart.js

Attended meetings with high-value clients to intake customer goals and receive feedback on designs

Created a tenfold increase in token count by rescraping 40 years of online content from marketing research journals using Python and Selenium WebDriver

Prevented request timeouts and decreased response times by 10 seconds through code structure improvement

Used convolutional neural networks to predict fault stability and induced seismicity for the safe disposal of nuclear waste

Implemented U-Net CNN and Mask R-CNN models to detect fractures and breakouts in borehole imagery

Achieved dice scores of 81.42% and 77.66% for U-Net and Mask R-CNN respectively, suggesting their ability to learn multiscale geologic features

Presented research findings to lab scientists and US Department of Energy member

Extended abstract in submission to CouFrac 2022