Senior Software Engineer(Ads Quality) at Coupang
Biography
Currently, I’m a graduate student for master degree in computer science(Machine Learning) at University of Southern California. As a senior software engineer for the past 6 years, I designed real-time/batch data pipelines and built recommendation and personalization systems, search relevance and ranking, ads targeting for e-commerce platforms. As a growth hacker, I collaborated with marketing, engineering and operation teams to analyze data and to perform offline experiments and online A/B testing to grow 17% of the user base and 20% of the daily revenue within 3 months. As a tech lead, I led global teams(Los Angeles, Beijing, and Shanghai) to develop and release the beta, v1 and v2 of Noomi Wallet, which supports Ethereum, Bitcoin and Cybermiles blockchain network.
- Ads Ranking
- Search Relevance & Ranking
- Machine Learning
- Recommendation & Personalization
- E-Commerce
-
M.S. in Computer Science, 2020
University of Southern California
-
BSc in Computer Science, 2012
University of Liverpool
-
BSc in Computer Science, 2012
Xi'an Jiaotong-Liverpool Univeristy
Skills
80%
80%
70%
70%
70%
70%
70%
90%
80%
95%
80%
80%
80%
60%
90%
90%
80%
80%
90%
80%
Experience
Responsibilities include:
- Developed new automated bidding algorithms and services to improve advertising platform efficiency.
- Collaborated with product and sales teams to design, implement and optimize new product ad features.
- Built large-scale machine learning models and systems to improve ad targeting, ranking, and relevance.
- Designed end-to-end monitoring workflow to track model inference, bidding, auction, and ranking quality.
- Provided suggestions and troubleshooting for advertisers to improve their return on investments.
Responsibilities include:
- Designed search ranking and personalization engine to improve relevance and engagement in search result page, home page, push notification and marketing campaigns.
- Utilized Convolutional Neural Network and Natural language processing techniques to recognize product images and generate product descriptions.
- Built end-to-end streaming and batch processing data pipeline to power feature engineering and model training.
- Mined million-level daily traffic to gain insights from user behavior and preference.
- Defined and implemented features, user profile, and segments on historical user behavior.
- Developed fraud detection tools to help operation team recognize fraud traffic and scam users.
Responsibilities include:
- Designed click-through rate model to improve page views and conversions on Women’s Clothing, Shoes and Bags, Wedding and Events, and Consumer Electronics categories.
- Built matrix factorization-based recommendation system to improve user’s engagement on product page.
- Utilized frequent pattern to find relevant buy-together products on checkout page.
- Maintained feature generation, model training, and deployment pipelines on AWS.
Responsibilities include:
- Maintained and scheduled mainframe workflow to process financial data.
Accomplishments
