Applied Scientist, Traffic Quality
Core
Build advanced capabilities at petabyte scale to detect sophisticated invalid traffic (IVT), including non-human traffic, bot networks, and fraudulent engagement patterns in programmatic advertising.
Role type
Applied Scientist II (fraud detection & traffic quality)
Builds
Production ML components and detection systems for Amazon Ads
Domain
Advertising technology, fraud detection, deep learning
Deliverable
production ML models
Required skills
deep learning, self-supervised learning, representation learning, advanced clustering, anomaly detection, time-series analysis, sparse labeling, algorithms and data structures, numerical optimization, parallel and distributed computing
Preferred skills
predictive modeling, large data analysis, experimental design, statistical analysis
Technologies
EC2, S3, EMR, Sagemaker, RedShift
Responsibilities
Define and frame new research problems in fraud detection; Invent and adapt new machine learning approaches to detect sophisticated invalid traffic; Design and deploy production-quality ML components; Apply domain knowledge to perform broad data analysis; Work with unstructured and massive datasets; Produce research reports meeting top-tier external publication standards; Mentor and develop junior scientists
Seniority
Mid-level IC (3+ years experience)
