Millions of Selling Partners trust Amazon's marketplace to grow their businesses, and hundreds of millions of customers depend on us every day. Behind that trust is a network of systems, tools, and workflows designed to detect and resolve fraud and abuse at scale — and our team builds them. We are looking for a Risk Manager - AI & Automation who will drive the intersection of risk intelligence and technology across TSI SCI and Res-Q. In this role, you will own defect elimination outcomes by leveraging AI-powered products and automation tools - identifying systemic abuse patterns, translating investigator insights into model improvement requirements, and measuring whether our technology investments actually reduce defect rates and reduce human investigation hours. You will partner with engineering, science, and operations teams to launch and optimize ML-based detection systems and automation workflows for risk reduction. You'll own root cause and preventive action (RC/PA) for automation-driven defect elimination, analyze ML model outputs to identify misclassification patterns, and reduce seller defect rates and improve investigator efficiency. This role offers the opportunity to shape how Amazon protects seller and customer trust by combining risk reduction expertise with technology execution that fundamentally improve how our organization operates.
Key job responsibilities
Own end-to-end defect elimination outcomes for AI and automation initiatives by coordinating across engineering, science, operations, legal, and policy teams to deliver on time and at quality
Drive cross-functional alignment across Risk Managers, Operations, Science, Legal, and Policy stakeholders to define product scope, prioritize features, and manage trade-offs - with defect elimination as the primary lens
Build and manage program roadmaps for automation launches, tracking dependencies, identifying risks, and driving escalation management across multi-team initiatives
Conduct defect pattern analysis using ML model outputs to identify systemic abuse trends, misclassification gaps, and opportunities for automation coverage expansion
Partner with science teams on ML model performance reviews, understanding precision/recall trade-offs, data pipeline health, and the operational impact of model threshold decisions
Translate investigator feedback into actionable model improvement requirements that serve as the voice of the risk operations team in technical prioritization discussions
Own root cause / preventive action (RC/PA) for automation-related defects - when models miss or misclassify, drive the investigation into why and translate findings into technical requirements
Own the RC/PA lifecycle for systemic defects surfaced through automation - when defect pattern analysis reveals recurring failure modes, document root cause findings, define preventive actions with measurable closure criteria, and track implementation through to validated defect elimination - ensuring accountability across science, engineering, and operations teams.
Don't want to miss the next one?
Subscribe to daily email alerts for roles matching your interests.
Lead escalation triage for automation-impacted workflows as the first-line risk assessor when automated systems produce unexpected outcomes (false enforcements, missed abuse signals, or seller experience degradation), rapidly diagnosing whether the issue is model drift, data pipeline failure, or policy misalignment, and routing to the appropriate team with a clear severity assessment and remediation path.
Conduct deep dives into escalation triage data, API integrations, and system gaps to identify launch blockers and ensure seamless product implementation
Define and instrument success metrics to measure automation coverage, defect escape rates, and seller experience impact
Prepare and present program updates, launch readiness reviews, and defect elimination impact assessments to senior leadership, quantifying the value of AI-driven risk reduction investments
A day in the life You'll start your day reviewing defect dashboards and ML model performance metrics - checking precision/recall trends, flagging misclassification spikes, and identifying where automation coverage is falling short. You'll meet with investigators and Risk Managers to understand which defect patterns are slipping through models and translate those insights into improvement tickets for science teams. You'll join system design reviews, refine technical specs for upcoming automation features, and drive sprint planning - prioritizing defect elimination impact. You'll analyze escalation triage data to spot systemic issues that technology should solve, and work with operations leaders to validate that deployed models are actually reducing investigation workload. Your week includes stakeholder alignment meetings with Science, Legal, Operations and Policy, launch readiness reviews, RC/PA deep dives on automation defects, and presenting defect elimination progress at WBRs and MBRs.
About the team
The Res-Q team operates as the central command for abuse-related escalations, conducting investigations and driving systemic improvements across Amazon's global stores. Through our various intake channels, we handle the most complex and sensitive cases that require expert judgment and cross-functional coordination.
### Basic qualifications
Bachelor's degree or equivalent
5+ years of program or project management experience
3+ years of experience working directly with engineering/science teams on product/feature launches
2+ years of experience in risk management, trust, fraud prevention, or marketplace integrity domains where the primary customer is an internal operations or investigations team
Experience managing cross-functional programs across technical and non-technical teams
Experience writing technical requirements, specifications, and design documents
Proficiency with SQL and understanding of data pipelines and data-driven decision making
Experience with defect analysis, root cause investigation, or process improvement in an operational setting
### Preferred qualifications
Experience launching AI/ML-powered products or automation tools that directly reduced operational defect rates or investigation workload
Hands-on experience with Amazon AI tools (Bedrock, SageMaker, or equivalent ML platforms)
Experience analyzing ML model outputs (precision/recall, confusion matrices) and translating performance gaps into actionable improvement requirements
Experience defining and tracking program metrics, building dashboards, and driving data-informed prioritization
Strong written and verbal communication skills to influence technical and non-technical audiences, including senior leadership
Familiarity with agile development methodologies and DMAIC process improvement frameworks
Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.
Similar roles you might like
More openings like this one — take a look before you go.
