Applied Scientist, Ad Matching & Large-Scale ML

Amazon; Greater London; On-site; GBP 100,000 - 150,000; Full time

Job summary

Amazon London is seeking an Applied Scientist to design and optimize ad matching systems, building deep learning models, evaluating them on large-scale datasets, and delivering features for programmatic advertising. Collaboration with cross-functional teams will be essential to turn research into scalable production solutions.

You will work with the Demand Retrieval team to push state-of-the-art ML methods, experiment rigorously, and help drive measurable improvements in campaign performance

Qualifications

• PhD, or a Master's degree and experience in CS, CE, ML or related field research.
• Experience programming in Java, C++, Python or related language.
• Experience in building machine learning models for business application.
• Experience in state-of-the-art deep learning models architecture design and deep learning training and optimization and model pruning.

Responsibilities

• Design and implement deep learning models to match users with ads across verticals/geographies.
• Investigate new ML techniques to support multiple advertisers/industries.
• Improve performance, generalisation and scalability of models with new features.
• Collaborate with engineers to deploy changes in large-scale ads stack.

Skills

Deep learning

ML model development

Data processing

Java

Python

Education

PhD in CS/CE/ML or related field

Master's degree in CS/CE/ML or related field

Tools

TensorFlow

PyTorch

Spark

Job description

Amazon London is seeking an Applied Scientist to design and optimize ad matching systems, building deep learning models, evaluating them on large-scale datasets, and delivering features for programmatic advertising. Collaboration with cross-functional teams will be essential to turn research into scalable production solutions.

You will work with the Demand Retrieval team to push state-of-the-art ML methods, experiment rigorously, and help drive measurable improvements in campaign performance