Background

My Story 👨‍💻

I grew up in Nanjing, China and Portland, Oregon 🌲. I love using data and code to tell stories, help journalist understand their readers better and enable newsrooms to grow and connect with their audience. I'm passionate about journalism, music, pottery, and reading.

UCLA

Statistics & Political Science

I double majored at UCLA 🐻, applying data science to analyze and catalyze social good.

Daily Bruin

Editor-in-Chief

I fell in love with journalism — reporter, editor, then EIC. I ran and won on a platform to improve diversity in the newsroom and its coverage.

UCLA VRP

Voting Rights Project

I used data science to support the center's legal team defending voting rights in Washington, New York, and Kansas.

Duke MIDS

Interdisciplinary Data Science

After two summers at H&R Block learning ML/DevOps, I came to Duke to go deeper on MLOps, Deep Learning, and Data Engineering.

Minnesota Star Tribune

Founding Data Scientist

I build bespoke ML and AI solutions to put great journalism in front of more readers and help the publisher navigate the digital media landscape.

Community

Asian American Journalists Association (AAJA) VP of Events

I organize the Minnesota chapter's annual trivia bowl and other events connecting Asian American journalists across the state.

Continuous Learning

News Product Management

I joined the News Product Management Certification program's fifth cohort to sharpen my product and leadership skills from industry leaders changing news with technology.

Clay

Pottery 🏺

I picked up pottery and love the meditative rhythm of the wheel. See what I've been making on my pottery studio page.


Work

Selected Projects

A mix of machine learning, data engineering, computer vision, and civic tech — built at Duke, UCLA, hackathons, and the Minnesota Star Tribune.

Arena Marketplace database project

Full-Stack · Databases

Arena Marketplace: Survival Gear Shopping for Tributes

A Hunger Games-inspired mini-Amazon built on a relational database. I integrated user accounts, product listings, and transactions, ensuring seamless functionality and a smooth shopping experience under simulated high-stress conditions.

PostgreSQL Python Flask Database Design
AI Image Captioning project

NLP · Vision-Language

Captioning AI-generated Images

Using BLIP and ViT models to reverse the text→image relationship. Our fine-tuned BLIP model showed a 25% improvement in average cosine similarity over the zero-shot baseline.

PyTorch BLIP ViT Fine-tuning
Breast Tumor Detection

Computer Vision · Healthcare

Detection of Breast Tumors with Simulated Motion Blurs

Used OpenCV to simulate motion artifacts and a custom Faster R-CNN model to localize tumors and determine which blur direction most impacts classification performance.

Python OpenCV Faster R-CNN PyTorch
Hidden Costs of Fashion app
3rd Place — HackDuke Environmental Track

NLP · Sustainability

Hidden Costs of Fashion

An image-to-text neural network that reveals the hidden electricity, water, and CO₂ costs of clothing items. Users can compare their purchases to the community and make more informed decisions.

Python NLP Neural Networks Heroku
FastAPI GoodReads application

Data Engineering · MLOps

Containerized FastAPI App with GoodReads Data

FastAPI application deployed via AWS ECR, CodeBuild, and App Runner, querying a books database. Full CI/CD pipeline with Docker containerization and automatic image pushes on commit.

FastAPI AWS Docker CI/CD Python
SQLite Twitter Demo

Data Engineering · SQL

SQLite Demo on Twitter Data

Used Python's Tweepy with SQLite to automate extraction of tweets from Joe Biden's timeline, then conducted data summarization and exploration through SQL queries.

Python SQLite Tweepy SQL
Global Temperature Prediction

Machine Learning · Climate

Global Temperature Prediction 🏖

Dask-parallelized time-series model across 3,000+ cities to forecast future temperatures and understand the disproportionate global impact of climate change. Deployed as a Streamlit app and CLI tool.

Dask Time-Series Streamlit Docker CI/CD
Cannabis and Drug Overdose Analysis
Best Visualization + ISI Florence Nightingale Finalist

Stats · Public Health

Cannabis Legality and Drug Overdose

UCLA Datafest award-winning project analyzing drug overdose risk for the Rocky Mountain Poison Center. A random forest classifier found recent cannabis use to be a major predictor of high overdose risk.

R Random Forest Data Visualization Public Health
Black Churches and Voter Turnout

Civic · Political Science

Black Churches and Voter Turnout

County-level analysis using ARDA data to understand whether the presence of predominantly African American churches strengthens Black voter turnout in the South.

R Regression Analysis Political Science Civic Data

Journalism

By Genesis Qu

Editor-in-Chief

Click on the slides below to read some of the stories I wrote and those that were written about me during my time at the Daily Bruin. As the first person of color to take charge of the newsroom in almost a decade, I was able to bring a new perspective to the newsroom and the stories we told. Photo credits to the incredible photographers at The Bruin, their names are in the stories.