Drimva

Amazon

Scaled Abuse Specialist, Special Projects & Investigations

$102K–$178K

Tempe, AZ +3 · Full-time · Posted May 2, 2026

Location
Tempe, AZ
Seattle, WA
Arlington, VA
San Diego, CA
Experience
3+ years of relevant industry experience; preferred qualifications also include experience in specified related fields.
Education
Bachelor's degree; a Master's degree in a specified field is preferred.

About the job

Investigates and prevents abuse and fraud across Amazon stores by analyzing data, identifying vulnerabilities, and scaling technical and policy solutions.

What you'll do

  • Analyze complex data and conduct large-scale investigations to identify abuse patterns, risks, and system or policy gaps.
  • Develop and scale detection methods and solutions, including using agentic AI to automate workflows and accelerate investigations.
  • Lead complex projects and stakeholder partnerships, communicating findings and recommendations to partner teams and executives.

What they ask for

  • Bachelor's degree.
  • 3+ years of relevant industry experience in risk or fraud investigations, Identity and Access Management, e-commerce, analytics, or security.
  • Proficiency deriving insights from big data using SQL, plus experience manipulating or processing data with Python and AI tools.
  • Proven ability to deliver complex projects across multiple teams.

Nice to have

  • Experience communicating with technical and non-technical audiences, including senior leadership.
  • Master's degree in Math, Statistics, Computer Science, or related Science field, or experience in data science, machine learning or data mining; experience with Amazon Redshift and other AWS technologies is also preferred.

Pay

102,000.00 - 178,400.00 USD annually (base salary), for each listed location

Tools

SQL, Python, R, Machine Learning, Data Science, Amazon Redshift, AWS technologies, agentic AI systems

AI focus

Uses agentic AI systems to automate workflows, scale pattern detection, and accelerate investigation and detection projects.

Model-written summary · the original posting is authoritative

First seen Oct 15, 2026 · Last verified Nov 6, 2026 (UTC)