Ronald Lee

May 2022: I'm excited to be joining Meta as a ML Engineer Intern on the App Ads ML team!

April 2022: Our work over crisis informatics was accepted into the International Journal of Information Management Data Insights!

March 2022: I joined the organizer team for TAMU Datathon, the first and only ML-focused hack-a-thon!

October 2021: At TAMU Datathon, we won 1st prize in Bloomberg's Industry Challenge!

August 2021: Our paper over network paradigm shifts affecting COVID-19 spread was accepted into Nature Scientific Reports!

TAMU Datathon

Director for Challenges and Logistics

Designing data science related challenges for contestants.

infolab

Undergraduate Researcher

Researching gradient balancing techniques in multi-task recommendation.

Meta (Facebook)

ML Engineer Intern

At Meta, I worked on the App Ads ML Team to prototype new multi-task learning models for Ads Recommendation. Specifically, I incorporated view-through data, which is logged when a user makes a conversion without clicking an ad, into our existing infrastructure. I landed the first data pipeline that used view-through data for our App Events models, and prototyped early findings that led to 0.03% NE gain. I also landed an API that automates the process of Cross-Domain Learning, where we mitigate signal loss via training on Android data and evaluating on iOS users.

Paycom

Software Developer Intern

At Paycom, I worked with the QA Team to develop a text based chat-bot with the capabilities to execute, analyze, queue, and edit regression tests. I implemented an NLP-based parser using ML.NET that could read in any text input and map keywords to the intended task, and perform that task efficiently. I also was able to use Paycom's internal API to implement the functionalities of the bot, such as execution and thorough analysis of tests. We used a C# backend for this project and a React Front-end component to integrate with Paycom's internal application. Furthermore, I also participated in a company hack-a-thon and used React to develop an interactive health leaderboard application.

UrbanResilience.AI Lab

Research Fellow

At the UrbanResilience.AI Lab, my research focuses on network science, where I analyze complex human interactions. One of my lasting contributions in the lab is creating a data processing pipeline that filtered billions of rows of anonymized geospatial data (coordinates, timestamps, ids, geometries, etc) into local contact networks for target counties. Through this, we were able to identify different interactions between anonymized users and calculate metrics to identify hidden risk in regard to the spread of COVID-19. I was also able to use Statistical and Machine Learning based models to help us uncover metrics that contributed heavily to pandemic spread, and it allowed us to guide local policy actions to mitigate this risk. You can read more about this in my publications linked in my google scholar!

Ansira Partners, Inc.

Human Resources Intern

At Ansira, I helped in the deployment “We University,” an internal learning library consisting of over 1,300 self-help courses.