Amazon
Applied Science Manager, Seller Partner Trust and Store Integrity Science - Global Risk Intelligence and Prevention
$184K–$249K
Seattle, WA · Manager · Full-time · Posted Oct 12, 2026
- Experience
- 3+ years of scientists or machine learning engineers management experience; also lists 1+ years of scientists or machine learning engineers management experience. Ph.D. in a relevant field, or PhD and 5+ years of experience in a quantitative field.
- Education
- Ph.D. in computer science, machine learning, engineering, or related fields, or PhD and 5+ years of experience in a quantitative field.
About the job
Leads a team developing and operating AI-powered systems to detect and prevent seller abuse across Amazon marketplaces.
What you'll do
- Set the scientific vision and roadmap, aligning priorities with stakeholders and communicating trade-offs.
- Lead scientific and technical work on detection models, ensuring rigor, scalability, and production readiness.
- Build and develop an applied science team and oversee project execution and operational performance.
What they ask for
- Knowledge of ML, NLP, Information Retrieval and Analytics.
- Ph.D. in computer science, machine learning, engineering, or a related field, or PhD and 5+ years of experience in a quantitative field.
- Experience hiring and leading experienced scientists and developing junior members.
- Experience with large-scale machine learning systems, including profiling, debugging, performance, and scalability.
Nice to have
- Experience building machine learning models or developing algorithms for business applications.
- Experience building complex software systems, especially involving deep learning, machine learning, and computer vision, delivered to customers.
Pay
USD 183,800.00 - 248,700.00 annually (base salary)
Tools
ML, NLP, Information Retrieval, Analytics, deep learning, foundation models, generative AI, large language models
AI focus
Lead development of AI/ML systems for seller risk detection, applying deep learning, foundation models, generative AI, large language models, multimodal reasoning, and self-supervised learning to identify abuse patterns.
Model-written summary · the original posting is authoritative