About
In 2018, as a rising junior in high school, I made the decision to enroll in a computer science course despite having no prior knowledge of coding. Fast-forward to today, and I’ve had the privilege of attending one of the top computer science programs in the nation. I’ve also gained experience building software for major departments on campus and have formed a strong network of supportive peers and mentors. In my free time, I enjoy listening to new music, playing 8 ball pool, hiking, and trying out new food spots.
Experience
Software Engineering Intern • Goplai
Fine-tuned custom YOLOv11 and SlowFast models for basketball player, ball, and hoop detections, manually annotating 500+ frames and clips to create high-quality datasets. Built a cloud-native, reat-time basketball analysis API with FastAPI, integrating detection, player tracking, scoring, and ball posession logic with AWS S3 storage and WebSocket streaming. Deployed application in Docker containers on AWS EC2 with automated health checks to ensure reliability and scalability.
AWSComputer VisionFastAPIDockerUltralytics YOLOREST APISoftware Developer • MBA Program for Executives, Haas School of Business
Developed a web application using Next.js that efficiently enables users to create personalized teams according to their unique needs. Build an approximation algorithm to generate optimally balanced teams based on user-defined priorities accross team size, gender, experience levels, and work history. Worked and collaborated alongside other departments on campus to expand the original scope of the project to encompass their teams and unique needs, including new criteria and data layouts.
ReactHTML/CSSNext.jsDanfo.jsAlgorithmsWebsite Developer • The Green Janitorial Corporation
Worked as a freelancer to create and develop a responsive website for The Green Janitorial using HTML/CSS, Next.js, and Bootstrap. Optimized the site for SEO and fast load times, improving search engine visibility and enhancing the user experience. Communicated and worked alongside the company owners to discuss project goals, steps, and timeline.
ReactBootstrapHTML/CSSSEOTech Fellow • UC Berkeley Career Center
Tested and implemented a Python script using BeautifulSoup and Selenium to scrape content from over 1000 employer websites to search for DEI Employment keywords. Also worked on updating employer contacts by parsing, analyzing, and visualizing data sets with over 200,000 data points to provide tangible and understandable solutions to an audience of team members with limited technical knowledge. Developed an interactive employee on-boarding web page using HTML/CSS, enhancing the user experience while streamlining the on-boarding process.
PythonPandasMatplotlibBeautifulSoupSeleniumHTML/CSSData Challenge Finalist • Meta
Explored, analyzed, and aggregated large data sets to provide actionable information, and create intuitive visualizations to convey those results to a broad audience. Also presented a data-driven product pitch, which included data visualizations, business strategy, and recommendations to Data Scientists and Data Engineers.
PythonSklearnPandasMatplotlibProjects
ICE Breakers
Developed a React Native mobile app that provides multilingual emergency support for immigrant communities, featuring a panic button, Red Card tool, and local resource integration.
React NativeTailwind CSSFirebaseExpress.jsSQL LiteREST APIRate My Classes
Built from the ground up using Next.js and MySQL, Rate My Classes is a web application that allows UC Berkeley students to rate and review their classes. The application uses an express.js server to handle API requests and a MySQL database to store user data such as login information, reviews, and ratings.
ReactHTML/CSSOAuthStyled ComponentsBootstrapSeleniumBeautifulSoupMySQLExposify
A full-stack web application that connects to the Spotify API and analyzes a user’s top artists and playlists to provide personalized music insights. Uses a backend using Next.js API routes to call the Spotify API and process data to generate musical profiles and analysis for logged-in users.
ReactHTML/CSSNext.js API RoutesREST APIVercelFacial Expressions Music Player
This web application uses facial recognition to detect a user’s facial expressions and play music based on the emotion detected. The application uses the face-api.js library to detect facial expressions.
JavascriptWeb SocketsFace-api.jsMaxMSPVibeS
Made during the 2023 CalHacks hackaton, VibeS is a web application that combines music and machine learning to create personalized playlists based on a picture. The application uses a Convolutional Neural Network to classify images based on their scenery and time of day to create a personalized playlist that fits the environment.
ReactNeural NetworksREST APIFull-Stack