Core Backend Development and Leadership

• Led agile backend development across three internal applications, eventually becoming the sole backend engineer on the Special Project Initiative (SPI) team within the first few months of joining the company and team

• Designed and implemented a universal security module (FastAPI & Django) adopted across many engineering teams

• Delivered bi-weekly API and architecture demos to the CTO, Head of Product, and revelent stakeholders

• Actively contributed to internal forums and cross-functional skill-ups in backend, frontend, and AI disciplines

• Spearheaded the first production release for three SPI applications respectively (Docker, AWS Lambda, & Jenkins)

Geolift, Mixed Marketing Models (MMM), Video Intelligence (VI)

• Constructed a restful DRF API to run complex Geolift marketing science models written by Meta in R and demoed functionality directly to Meta clients

• Developed a queuing system for AWS Batch jobs to run complex functions asynchronously, now used by internal teams

• Pioneered the first client application (MMM) implementation of Trino and FastAPI, setting precedent for future use

• Re-architected an existing Laravel application (VI) into a modern stack (FastAPI, Lit) to adapt and analyze YouTube API's

• Contributed to full stack development (Lit, JS, TS), such as building pagination, polling, and design enhancements

• Led 20+ students to construct and design a new, 38 home, EcoVillage in Charlottesville

• Co-led frequent meetings with the EcoVillage's project manager, and bi-weekly team meetings to create educational designs that teach "Common House" visitors on resource usage, renewable/sustainable practices, and eco-friendliness

• Taught first-year students and new club members to create sustainable architecture via multiple AutoCAD programs

• Core contributor to the "Buildings on Grounds" committee, which propelled marking enslaved laborers burial sites with respectful signage and acknowledgment to UVA history, as well as added ramps to inaccessible living spaces (UVASC)

• One of 50 elected students who represented the entire 4,200+ class of 2023, designed to support and improve the daily life and activity for these students (Second, Third, & Fourth Year Council)

• Co-led the "Dinner Series", "Third Year Ceremony", and "Big Events", committee to invite, and host, internationally renowned speakers to speak to and celebrate the entire class of 2023 (Second, Third, & Fourth Year Council)

• Organized event-fundraising efforts; cultivated for nearly 700 first-year engineering students (Eng. Student Council)

• Led the creation of a core-feature Azure function application (C#) and 28 neural network machine learning models, leveraging agile development practices to deliver in weekly sprints

• Polished the application in order to fulfill its purpose in analyzing unique tickets for industrial wood transportation and sales, which is a landmark feature for Legna's SaaS platform for the forestry industry

• Designed the application to only be triggered through event grid subscription; ensuring it is only called when images were uploaded to blob storage containers (multi-tenant; specific to each customer)

• Curated nearly 3,000 images stored in an Azure storage container for model training

• Authored an Azure function app triggered by an HTTP request used internally to fill data for nearly 70 demos per year

• Revolutionized the forestry ticket system industry by being one of, if not the first, people to introduce AI recognition

• Created and maintained software to generate heatmaps for degree, closeness, “betweenness”, eigenvector centrality, and MRSA patient count for the 8-floor, 651 bed, UVA Hospital

• Architected a pipeline to ingest data from 2 high powered computing clusters (with over 595 nodes and 22598 CPU cores) and then processing that data to assign correlated, positive MRSA cases based on each room

• Leveraged yEd, graphML, and NetworkX to create an accurate network representation of the hospital

• Implemented over 10 Jupyter Labs notebooks to leverage the network’s tree-style structure to retrieve information regarding neighboring rooms based on a given room number