• Serving as main developer and subject matter expert for the Data Interface Governance tool, an internal web application to help manage metadata for automated data file processing. The tool is currently released in production, and is predicted to speed up ingestion for onboarding new data providers by 6-12 months compared to the previous procedures.

o Utilized Python FastAPI for backend API, and Angular for frontend web UI

o Incorporated NLP model to automatically identify and mark record fields that could contain Personally Identifiable Information, with 90% accuracy

I have also worked on a Reverse ETL data pipeline to facilitate migration from on-premises databases to cloud database (Snowflake), utilizing Airflow.

• Researched benefits and drawbacks of quantum computing in the application of image processing-related Machine Learning Models (Generative Adversarial Networks)

• Implemented CycleGAN architecture with classical and quantum components and compared/analyzed training time and performance

• Delivered final presentation & analysis to team to inform decisions on future projects

I worked working part-time as a teaching assistant for Java Labs at Columbia University. Java Labs are weekly 2-hour sessions for students who want to reinforce the concepts they learn in their Introduction to Java class.

Each session, after a review lecture, I lead a group of students, starting them on the project for the week. I then answer any questions they have as necessary, guiding them towards solutions and helping them build a strong understanding of the material.

In the summer of 2021, I worked as an Undergraduate Researcher in the NSF Immersive Media Computing REU (Research Experience for Undergraduates) at Georgia State University. For 8 weeks, we worked online with professors and graduate student advisors on research projects. Alongside research, we also attended several workshops in areas such as game development and computer vision, learning about new technologies & research in the immersive media fields.

Along with another undergraduate student, I worked on optimizing an existing video inpainting model to work on 360º video in real-time. These deep learning models utilize spatial & temporal redudancy in videos to fill in missing or masked regions in video, but are usually very computationally intensive. We aimed to decrease the computation of the model (and increase the speed) with minimal impact on the inpainting quality.

I was one of 10 students selected for this program, out of 292 applicants nationwide.

Served as a teaching assistant for the course Beyond Bits & Atoms for Spring 2020 & Spring 2021(http://beyondbitsandatoms.org/pages/course-info.html) at Columbia University's Teacher's College. Students were tasked with learning constructionist education concepts and applying them to create educational technologies for students. Throughout the semester, I assisted the students in various aspects of their design, prototyping, and development of their projects, including the transition online in the middle of the semester. I focused mostly on the incorporation of technology, and led workshops on Arduino and basic computer vision.