
Computer Science
UCLA '21
Hi! I'm David and I'm currently a software engineer at Google. I majored in computer science at the University of California - Los Angeles. During my time at UCLA, I've gained a keen interest specifically in machine learning and utilizing data in order to build models with real-life applications.
My ultimate goal is to have (positive) stuff about me show up on a Google search of my name. Stay tuned to check out what I've been up to!
Maps Search and Suggest Ads
Created and visualized key metrics to identify potential bottlenecks within the Search Web Rendering Service (SWRS) page view hierarchy.
Refactored and deployed candidates under experimental weblabs to reduce DOM tree node depth. Changes benefitted customers in all locales (worldwide) on every retail page search query.
Built out more robust tooling to automatically identify critical loading sections on the page. Integrated Google Lighthouse into the team's pipelines to provide workflow analysis.
Created automatic guard rails for Amazon UI packages to ensure quality customer experience and protect against deploying sub-optimal changes.
Enhanced the internal Sled tool to work seamlessly with guard rails for greater developmental agility.
Integrated functionality with credential management and trouble ticketing system to validate Partner contact information. Automated and streamlined diagnosis and issue-handling processes.
Developed an extendable and flexible test automation framework for database interaction with various GUI data visualization tools. Utilized in Jenkins continuous integration testing and proof-of-concept.
Designed custom image-matching with user-specified precision using cv2 to ensure simple transitions between different resolution monitors.
Here are a couple personal side projects I've been working on in my free time...
Created a custom Selenium/Beautiful Soup social media web scraper. Authenticates via log-in or local cookie cache and pulls in data from a configurable amount of available posts. Adding topic model clustering for text analysis capability and RNNs (for each cluster learned) for realistic text generation. Maybe this will be consolidated into an automatically generated report?
Coded a neural network from scratch in C++, including both forward and backward propagation. Python is definitely better from a practical sense, but I thought it would be fun to test my own understanding and challenge myself (no fancy libraries or matrices used)!
Used a mixture of numerical, categorical, and text data to determine substitutes for Amazon products to display to customers. Investigated a couple models, including KNN, decision trees/random forests, and neural networks using Amazon SageMaker.