Machine Learning Engineer, Recommendation - E-Commerce

About the Teams E-commerce - Recommendation and Marketing E-commerce is a new and fast growing business that aims at connecting all customers to excellent sellers and quality products on TikTok Shop, through E-commerce live-streaming, E-commerce short videos, and commodity recommendation. We are a group of applied machine learning engineers and research scientists that focus on E-commerce recommendations. We are responsible for building up and scaling our recommendation system to provide the most stable and best shopping experience for our TikTok users.

E-commerce - Alliance The E-commerce Alliance team aims to serve merchants and creators in the e-commerce platform to meet merchants' business indicators and improve creators' creative efficiency. By cooperating with merchants and creators, we aim to provide high-quality content and a personalized shopping experience for TikTok users, create efficient shopping tools at seller centers, and promote cooperation between merchants and creators.

E-commerce - Search The Search E-Commerce team is responsible for the search algorithm for TikTok's rapidly growing global e-commerce business. We use state-of-the-art large-scale machine learning technology, the cutting-edge NLP, CV and multi-modal technology to build the industry's top-class search engine to provide the best e-commerce search experience, for more than 1 billion monthly active TikTok users around the world. Our mission is to build a world where "there is no hard-to-sell good-priced product in the world".

E-commerce - Search Growth The Search Growth E-commerce team is at the forefront of developing the search recommendation algorithm for TikTok's rapidly expanding global e-commerce enterprise. Utilizing cutting-edge machine learning technology, advanced NLP, CV, recommendation, and multi-modal technology, we're shaping a pioneering engine within the industry. Our objective is to deliver the ultimate e-commerce search experience to over 1 billion active TikTok users worldwide. Our mission: to create a world where "there are no hard-to-sell, overpriced products."

What You Will Do

• Responsible for the build and design of optimization algorithm strategies for large-scale (10 million to 100 million products or creators' contents) e-commerce recommendation algorithm pipeline
• Build long and short term user interest models, analyze and extract relevant information from large amounts of various data and design algorithms to explore users' latent interests efficiently.
• Design, develop, evaluate and iterate on predictive models for candidate generation and ranking(eg. Click Through Rate and Conversion Rate prediction) , including, but not limited to building real-time data pipelines, feature engineering, model optimization and innovation.
• Design and build supporting/debugging tools as needed.

Minimum Qualifications

• Bachelor above degree in computer science or relevant areas.
• 3+ years of experience with a solid foundation in data structure and algorithm design, and be proficient in using one of the programming languages such as Python, Java, C++, R, etc.;
• Familiar with common machine/deep learning, causal inference, and operational optimization algorithms, including classification, regression, clustering methods, as well as mathematical programming and heuristic algorithms;
• Familiar with at least one framework of TensorFlow / PyTorch / MXNet and its training and deployment details,as well as the training acceleration methods such as mixed precision training and distributed training;
• Familiar with big data related frameworks and application, those who are familiar with MR or Spark are preferred

Preferred Qualifications:

• Experience in recommendation systems, online advertising, ranking, search, information retrieval, natural language processing, machine learning, large-scale data mining, or related fields.
• Publications at KDD, NeurlPS, WWW, SIGIR, WSDM, ICML, IJCAI, AAAI, RECSYS and related conferences/journals, or experience in data mining/machine learning competitions such as Kaggle/KDD-cup etc.