Machine Learning Engineer
CurrentPromotion Hub: Voucher Ranking 🎰 Implemented real-time Multi-Armed Bandits (MAB) framework to optimize the promotion recommender system at scale for a user base of 40 million, utilizing Bisecting K-Means clustering model and Thompson Sampling algorithm. 📈 Conducted rigorous A/B tests, leading to an impressive 18% increase in click-through rate (CTR) and a 3% boost in user engagement, demonstrating the effectiveness of the implemented MAB strategies.⚡️ Developed and maintained end-to-end data pipelines, encompassing data collection, cleaning, and preparation at scale, while ensuring seamless integration with real-time data streams from Kafka, and real-time serving with HBase.