Applied Scientist, Amazon Transportation
Amazon · London, United Kingdom · Applied Science
About this role
Amazon is hiring a mid-level Applied Scientist in the machine learning function based in London, United Kingdom. The posting calls out experience with Python, Java, Networking, Data Structures.
- Role
- Applied Scientist
- Function
- machine learning
- Level
- mid
- Track
- Individual contributor
- Employment
- Full-time
- Location
- London, United Kingdom
- Department
- Applied Science
- Posted
- Mar 30, 2026
More roles at Amazon
Job description
from Amazon careersAmazon's Middle Mile Science group is looking for an Applied Scientist to build machine learning and optimization models for large-scale transportation planning systems. This includes the development of dynamic pricing and network planning models to improve operations and services for our external freight customers. The Middle Mile Science group develops optimization and machine learning systems that power Amazon's freight transportation network, from network design and pricing to real-time load planning and capacity utilization. The scale of Amazon's fulfillment operations requires robust transportation networks that minimize cost while meeting all customer deadlines. Real-time execution depends on state-of-the-art optimization and artificial intelligence to coordinate thousands of operators and drivers. This includes shipper-facing and carrier-facing marketplace algorithms as well as network planning and optimization tools. Amazon often finds that existing techniques do not match our unique business needs, driving the innovation of new approaches and algorithms. As an Applied Scientist focusing on external freight within middle mile transportation, you will work closely with business leaders, engineers, and fellow scientists to design and build scalable products operating across multiple transportation modes. You will create experiments and prototypes of new machine learning and optimization applications, present research findings to senior leadership, and implement your models…