Senior Machine Learning Engineer - TikTok E-commerce Global Supply Chain and Logistics
TikTok.com
Office
Seattle, Washington, United States
Full Time
The E-commerce Global Supply Chain and Logistics team is dedicated to enhancing clients' shopping experience and reducing logistics operational cost in TikTok E-commerce. We are currently looking for talented software engineers that have a deep understanding of machine learning (ML), operations research (OR), data mining and statistical inference. This position can be fulfilled in our San Jose and Seattle offices.
Responsibilities
- Build deep learning and statistical models to provide end-to-end estimated time of arrival (ETA) prediction for the e-commerce logistics. Through fine adjustment of ETA expression to the consumers, improve the click-to-order rate while balancing the negative review indicators.
- Use data mining tools to build logistics network knowledge graphs, based on which to construct a situational awareness & early warning system for logistics fulfillment, enabling operations to discover and deal with logistics network anomalies, and to improve the logistics fulfillment quality together with logistics service providers.
- Analyze and predict the spatio-temporal trajectory sequence of express packages through deep learning, statistical inference and other algorithmic methods. This trajectory prediction can help to build a better understanding of logistics network dynamics, improve ETA prediction accuracy and provide important sample features for other prediction tasks.
- Build a service network design (SND) model for the location selection of transshipment centers and last mile stations based on an expected increase in logistics order volume.
- Extract the direction of logistics operation optimization, cost reduction and service quality improvement by in-depth understanding of supply chain and logistics scenarios.
- Develop innovative and state-of-the-art e-commerce logistics models and algorithms
Responsibilities
- Build deep learning and statistical models to provide end-to-end estimated time of arrival (ETA) prediction for the e-commerce logistics. Through fine adjustment of ETA expression to the consumers, improve the click-to-order rate while balancing the negative review indicators.
- Use data mining tools to build logistics network knowledge graphs, based on which to construct a situational awareness & early warning system for logistics fulfillment, enabling operations to discover and deal with logistics network anomalies, and to improve the logistics fulfillment quality together with logistics service providers.
- Analyze and predict the spatio-temporal trajectory sequence of express packages through deep learning, statistical inference and other algorithmic methods. This trajectory prediction can help to build a better understanding of logistics network dynamics, improve ETA prediction accuracy and provide important sample features for other prediction tasks.
- Build a service network design (SND) model for the location selection of transshipment centers and last mile stations based on an expected increase in logistics order volume.
- Extract the direction of logistics operation optimization, cost reduction and service quality improvement by in-depth understanding of supply chain and logistics scenarios.
- Develop innovative and state-of-the-art e-commerce logistics models and algorithms