Developing a lightweight user representation model for Daum, a popular news portal service with 10 million users, by leveraging important keywords from user-read articles to capture evolving user interests. The output of model served as a context input feature for bandit models, enhancing news recommendation quality.

Building personalized recommender systems for mm, an audio-only social network services like Clubhouse. I employed graph-based recommendation models, incorporating techniques like graph pruning and regularization to optimize performance.

Combining content-based representation learning and collaborative filtering to address cold-start issues for new comics on Kakao Webtoon and Piccoma in Korea and Japan. The recommender systems served as a user-targeted marketing tools, optimizing the first conversion rate.

Creating a context-aware recommender system for Kakaotalk Gift, an e-commerce platform for sending gifts throught Kakaotalk messenger. The system considered multiple contexts such as demographics, product popularity, and user history including clicked and purchased items, by integrating text and image embedding along with collaborative filtering techniques.

Developing and maintaining machine learning applications for shoppinghow, an e-commerce platform similar to eBay and Amazon. I implemented product categorization and matching systems using Transformers and graph algorithms, applying parallel computing techniques to handling billion-scale data efficiently.