I am currently a senior undergrad student studying at UC San
Diego pursuing a degree in data science and a minor in computer engineering. I enjoy
working out, hanging out with
friends, and playing video games.
📄 Read My
Résumé Here!
GPA: 3.77
Relevant Coursework: Machine Learning, Deep Learning, Natural Language Processing, Data Analysis, Computer Organization and Systems Programming, Data Structures & Object-Oriented Design, Differential Equations, Linear Algebra, Probability and Statistics, Discrete Mathematics, IoT & Sensors, Components and Design Techniques for Digital Systems
Developed a custom convolutional neural network with PyTorch to classify 20 categories of food images. Constructed and trained the neural net with PyTorch and cuda. Optimized the network through changing the architecture and the hyperparameters. Graphed and analyzed the feature maps and the filters with matplotlib.
Developed a configurable LSTM model to classify the sentiment of Twitter tweets into different categories and created an API that returns a response JSON object of the sentiment result with FastAPI. Trained 1000 different LSTM models with grid search to find the optimal hyperparameters.
Created a Python script that downloads hottest videos from subreddits with the Reddit API and uploads them to tiktok accounts with Python and Selenium. Can be deployed to a server and post videos automatically with configurable settings (rest interval, target subreddit, etc.)
Developed Minecraft plugins that improves/expands the player experience by adding items to interact with in Minecraft. The most popular plugin current has over 2.8M downloads on CurseForge.
Created three data visualization with Python and brief explanation of their importance in the development of data science.
Developed a touch-based air hockey game with C# under the Unity game engine with a functional alpha version with a proper game mechanism, scoring system, UI animations, and an AI trained through reinforcement learning with Unity’s ML-Agents library Uses C# scripts to create custom behaviors and trigger events. Uses Blender to model the game objects. Uses Photoshop to adjust the UI sprites.
Created a configurable Python webscraper that checks Raspberry Pi stocks from verified sellers and prints out the information in colored and formatted texts. Uses request to download the html source code of each webpage and BeautifulSoup4 to load the html code and find the tags containing the desired information. Uses termcolor to format the outputs.