Job Description

: Collaborate closely with the Shopee promotion and campaign product team, business partners, and internal stakeholders to identify market growth challenges, extract insights, pinpoint key drivers, and deliver practical solutions based on extensive user and product data. Construct fundamental machine learning models that encompass deep learning and operations research, while also establishing end-to-end machine learning pipelines to enhance the efficiency of promotion products and enhance critical performance metrics like CTR and CR. Create algorithms for the extraction, transformation, and loading of large volumes of batch and real-time data, structured and unstructured, to implement AI solutions for addressing business issues. Conduct experiments to evaluate performance and address any issues encountered during the process. Engage in projects related to personalized recommendations and the optimization of voucher, banner, and promotion item distribution, intelligent recommendations for voucher and promotion item configuration settings, e-commerce campaign budget planning, product demand forecasting, and more. Requirements: Hold a Master\'s degree or higher in fields such as computer science, information technology, data mining, or computational statistics. Possess a minimum of 3 years of cumulative work experience as a data scientist or AI/algo engineer in large e-commerce platforms. Proficiency in Python and SQL programming is mandatory. Having substantial experience with machine learning frameworks such as Tensorflow, Keras, and Pytorch, as well as expertise in big data analysis and distributed systems like Hadoop, Hive, and Spark is mandatory. Capable of analyzing and synthesizing data from both quantitative and qualitative sources to derive insights for designing practical and scalable solutions. Exhibit strong logical analysis skills, a thirst for learning, and a collaborative team spirit. Possess a solid track record in natural language processing and multimodal learning algorithms (e.g., Seq2seq, CLIP, Prompting, LLM, etc.) is required. Familiarity with recommendation system algorithm architecture and processes, including experience with recall, ranking, re-ranking deep learning models (e.g., DIN, DIEN, MMOE, PRM, Multi-Domain models, etc.). Must have at least one published paper in top-tier conferences or journals within the fields of recommendation systems or natural language processing, such as KDD, ICDE, Recsys, ACL, EMNLP, ICMI.

foundit