Applied Scientist, Global Risk Intelligence & Prevention

Amazon.com Services LLC; Seattle (WA); On-site; USD 143,000 - 193,000; Full time

Benefits offered by this job

Health insurance

Restricted stock units (RSUs)

401(k) matching

Paid time off

Parental leave

Job summary

Amazon.com Services LLC is seeking an Applied Scientist for Seller Abuse Prevention to lead AI-powered risk detection and abuse prevention across the seller lifecycle. You will design, develop, and deploy scalable AI solutions using NLP, LLMs, and multi-modal data to detect bad actors at scale.

You will own end-to-end scientific work from risk quantification to decision optimization, and you will build interpretable pipelines that justify model decisions for enforcement and seller experience.

Qualifications

• PhD or Master’s degree in CS/CE/ML or related field with 4+ years of experience.
• Strong programming experience in Java, C++, Python or related language.
• Experience with algorithms, data structures, numerical optimization and large-scale systems.

Responsibilities

• Design and build predictive risk detection models using advanced AI techniques including NLP and LLMs.
• Own end-to-end scientific solutions from risk quantification to decision actions.
• Develop interpretable pipelines explaining model decisions for enforcement and seller experience.
• Collaborate with risk programs and engineering to deploy models in production and measure impact.
• Identify new abuse vectors and translate patterns into automated risk-detection systems.

Skills

Java

C++

Python

Algorithms/Data Structures

Numerical Optimization

Education

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

Tools

Unix/Linux

Job description

Amazon.com Services LLC is seeking an Applied Scientist for Seller Abuse Prevention to lead AI-powered risk detection and abuse prevention across the seller lifecycle. You will design, develop, and deploy scalable AI solutions using NLP, LLMs, and multi-modal data to detect bad actors at scale.

You will own end-to-end scientific work from risk quantification to decision optimization, and you will build interpretable pipelines that justify model decisions for enforcement and seller experience.