💼 Software Engineer Intern @ Con Edison (Jan 2023 – Present)

New York, NY | Python, Java, SQL, LangChain, Outlook API, Flask, PostgreSQL, cron

During my time at Con Edison, I drove key automation initiatives that streamlined internal operations and reduced manual overhead. Highlights include:

🔄 Mass Update Management System

Built backend components for a scalable contact management system and dashboard that tracks the status of thousands of EMOPSYS customer accounts during the bi-annual four-month update cycle.

🤖 LLM-Powered Mailing Agent

Developed an intelligent email automation pipeline using LangChain and the Outlook API to send personalized contact update requests to customers, automatically parse responses, and update the database.

Scheduled and orchestrated monthly email waves with cron, ensuring timely outreach while dynamically removing responders to prevent duplicate messages.

Significantly improved communication efficiency and customer engagement through automation and NLP.

🐍 Automated Data Pipelines

Designed and implemented Python-based automation scripts for internal data processing, eliminating repetitive workflows and saving the department over 450 hours of manual labor annually.

As a dedicated Computer Science Instructor at the AI Plus Learning Center, a dynamic cram school for young students eager to delve into the world of computer science, I played a pivotal role in shaping the next generation of tech enthusiasts. Key contributions include:

🖥️ Lab Session Facilitation:

Conducted engaging lab sessions to bridge theoretical concepts with hands-on applications, empowering students to solve practical problems in the realm of computer science.

💻 Coding Guidance and Support:

Provided personalized guidance on coding assignments, offering assistance in debugging code and troubleshooting errors. Ensured students developed a robust understanding of programming principles.

👩‍🏫 Programming Languages Instruction:

Led classes on Python, Netlogo, Java, and Object-Oriented Programming, fostering a comprehensive understanding of diverse programming languages and paradigms.

At EmmyCo, a company revolutionizing patient care, I played a crucial role in ensuring the success of our mission to offer an Uber-like service, where we transported patients from hospitals to their homes when familial support was unavailable.

Within this dynamic startup environment, I gained invaluable experience in diverse facets of software development and healthcare technology, notably excelling in:

🔄 Agile Development Practices:

Embraced agile development methodologies to foster adaptability and responsiveness. Engaged in regular sprint planning, retrospectives, and daily stand-ups to enhance communication within the team and facilitate a rapid and iterative development cycle.

📈 Performance Optimization:

Utilized performance testing tools such as JMeter and Selenium to identify and address bottlenecks in the system, ensuring a seamless and efficient experience for users on our company website. Collaborated closely with the development team to implement optimizations and enhance overall platform performance.

🛡️ Backend Unit & Integration Testing

Improved software quality by implementing thorough unit and integration tests and identifying critical defects increasing code coverage by 30%.

During my research internship at the University of Florida, I shadowed Dr. James Hamlin who does research on the synthesis of superconductive compounds under high pressure. Superconductivity, characterized by zero resistance at cryogenic temperatures, holds immense potential, particularly in creating power lines with no transmission loss. Yet, the impractical refrigeration costs have spurred a significant materials science effort to discover room-temperature superconductivity for more practical large-scale applications. My specific contributions included:

🔍 Modeling Superconductivity:

Presently, the search for room-temperature superconductivity often relies on an inefficient trial-and-error process. A primary objective of my internship was to develop a regression and classification model that could predict superconductivity in compounds and the critical temperatures at which they become superconductive. The motivation behind this was to provide researchers with a valuable tool—a preliminary filter to markedly reduce the pool of compounds requiring testing. For my regression model, I had used and implemented SISSO, a feature construction and selection algorithm originally developed in Fortran, in Python. Ultimately, my regression model didn't do as well as I had hoped but my classification model achieved an accuracy of roughly 73% which was surprisingly high given the size of the dataset I had to train on.

📊 Data Curation:

Curated a dataset from our database of superconductive compounds which was used to train the machine-learning models.