Managing a team of Data Scientists, Machine learning Engineers, Business insight analysts, Front-End developers, and QA to deliver different recommendation products including content/product/offer personalization across Web/App/Email channels. Squad focuses on Research and Development produce on Personalisation recommendations and using cutting-edge NLP approach to search for personalization.* Large-Scale Real-Time Personalisation RecommendationDesigned and developed a state-of-the-art real-time product recommendation engine, which dynamically recommends products and recipes based on their purchase/online behaviors to create profit and engagement generating recommendations to the customer. The project utilized various technologies such as Transformer, Graph Neural Network, and other large-scale deep learning architecture.* NLP enhanced Search OptimizationMultiple data mining and NLP techniques were utilized to enhance the customer search experience, such as NLP keyword analysis, NLP filters & synonyms, and another personalisation capability in ML for search * Real-time and model serving Real-time capability development in feature serving (low latency feature engineering and processing) and model serving (serving with End-points and traffic controller to design real-time AB testing and online retrain method)* MeasurementMeasurement design, create business focus metric for model measurement and how to capture and leverage this to improve model quality to make sure zero lag update.

Managing a team of Data scientists, Machine learning Engineers to cover a wide range of POC and business deliveries. * Search PersonalizationBuilding Next Basket Recommendation Model to enhance the personalized search* Recipe Content Personalization Building different recommendation strategies for recipe content personalization by coinciding with customers’ online behaviors. The key strategies include trending, popular, and personalization. Built fully automated end-to-end pipeline from data extraction to modeling and real-time serving.* Real-Time Recommendation ArchitectureLeading a team of data scientists exploring state-of-the-art recommendation algorithms such as transformer-based sequential recommendation, and graph neural network-based recommendation.

Developing in-house Automotive Machine learning algorithms and platforms for distributed training of massive data in a cost-effective way. Building personalised decisions engineer to provide the right customers with the right offer at the right time.Tool involve: Google cloud Platform/Pub-sub/Argo, Spark, Python, R, Scala, AWS platform, and numerous machine learning packages.

• Preparing tutorial notes, tutoring master students Machine Learning course

• Short-term/Long-term prediction of Train operation• Development of data-driven train timetable evaluation model

• Preparing tutorial notes• Tutoring master students Machine Learning course

I have developed an ultrafast clustering method with 50x speedup against state-of-the-art algorithm and written two first author peer-viewed publications.