Senior Software Development Engineer, Amazon Publisher Monetization (APM) Insights
Summary
Senior engineer building and scaling an AI-powered insights platform that turns publisher monetization data into fast, reliable, and self-service insights for Amazon’s ad teams.
We're looking for a Senior Software Development Engineer (SDE) to build and scale APM's AI-powered insights platform which supports 22+ 1P publisher monetization businesses. Our goal is to minimize the data-to-insight cycle time across both structured and unstructured data, and provide AI-powered Insights-as-a-service. You will own the end-to-end technical strategy and delivery of production-grade AI services.
Our team includes data and business intelligence engineers who bring deep domain expertise in data analytics and publisher monetization. They own the structured data foundations including data models, aggregates, and certified metrics. You will partner closely with them to build the application and service layer, ensuring that our AI-powered experiences are grounded in high-quality, governed data assets. Success in this role requires strong collaboration with DEs and BIEs.
You should be energized by building 0-to-1 while maintaining the rigor of a tier-2 service owner. You will navigate ambiguity, make trade-off decisions independently, and influence the technical direction of the platform without waiting for direction. You will be expected to raise the engineering bar for the team through your technical judgment and operational excellence.
Key job responsibilities
- Own the architecture, reliability, and evolution of the AI insights platform end-to-end, making high-judgment trade-off decisions between speed, quality, and scope
- Design for multi-tenant consumption: build systems that serve multiple APM teams reliably without bespoke integrations per consumer
- Establish tier-2 operational guardrails: monitoring, alarming, automated rollback, load testing, dependency management, and on-call runbooks
- Own service-level objectives (SLOs) and drive accountability for uptime and latency; design and implement AI safety mechanisms: output validation, hallucination detection, and confidence scores
- Build self-service onboarding and configuration systems that allow new APM teams to consume insights capabilities without custom engineering work; instrument usage telemetry to inform platform evolution
- Co-define the contracts between structured data foundations, unstructured business context, and the application layer, minimizing the need for manual intervention in updating AI experiences
- Partner with BIEs to build relevant user experiences along with evaluation frameworks that validate insight accuracy, relevance, and business applicability before AI outputs reach end users
- Define software engineering standards (code review practices, testing requirements, deployment processes, documentation norms)
About the team
Amazon Publisher Monetization (APM) is responsible for increasing ad monetization across 22+ 1P publishers (like Prime Video, Twitch, Alexa, Stores) by delivering compelling ad experiences while safeguarding CX. The APM Agentic Insights team builds scalable intelligence that delivers fast, accurate, and actionable insights across APM teams. We serve business, product, and science teams, replacing manual intake-driven analytics with self-service, AI-powered experiences. Our solutions range from off-the-shelf queries and dynamic dashboards to conversational analytics agents and MCPs. Our vision is to be the trusted insights partner for every APM business, driving measurable business impact.
Our team includes data and business intelligence engineers who bring deep domain expertise in data analytics and publisher monetization. They own the structured data foundations including data models, aggregates, and certified metrics. You will partner closely with them to build the application and service layer, ensuring that our AI-powered experiences are grounded in high-quality, governed data assets. Success in this role requires strong collaboration with DEs and BIEs.
You should be energized by building 0-to-1 while maintaining the rigor of a tier-2 service owner. You will navigate ambiguity, make trade-off decisions independently, and influence the technical direction of the platform without waiting for direction. You will be expected to raise the engineering bar for the team through your technical judgment and operational excellence.
Key job responsibilities
- Own the architecture, reliability, and evolution of the AI insights platform end-to-end, making high-judgment trade-off decisions between speed, quality, and scope
- Design for multi-tenant consumption: build systems that serve multiple APM teams reliably without bespoke integrations per consumer
- Establish tier-2 operational guardrails: monitoring, alarming, automated rollback, load testing, dependency management, and on-call runbooks
- Own service-level objectives (SLOs) and drive accountability for uptime and latency; design and implement AI safety mechanisms: output validation, hallucination detection, and confidence scores
- Build self-service onboarding and configuration systems that allow new APM teams to consume insights capabilities without custom engineering work; instrument usage telemetry to inform platform evolution
- Co-define the contracts between structured data foundations, unstructured business context, and the application layer, minimizing the need for manual intervention in updating AI experiences
- Partner with BIEs to build relevant user experiences along with evaluation frameworks that validate insight accuracy, relevance, and business applicability before AI outputs reach end users
- Define software engineering standards (code review practices, testing requirements, deployment processes, documentation norms)
About the team
Amazon Publisher Monetization (APM) is responsible for increasing ad monetization across 22+ 1P publishers (like Prime Video, Twitch, Alexa, Stores) by delivering compelling ad experiences while safeguarding CX. The APM Agentic Insights team builds scalable intelligence that delivers fast, accurate, and actionable insights across APM teams. We serve business, product, and science teams, replacing manual intake-driven analytics with self-service, AI-powered experiences. Our solutions range from off-the-shelf queries and dynamic dashboards to conversational analytics agents and MCPs. Our vision is to be the trusted insights partner for every APM business, driving measurable business impact.