Software Dev Engineer II, Sales AI

Are you interested in shaping the future of Advertising and B2B Sales? Amazon Advertising is one of Amazon's fastest growing and most profitable businesses, delivering billions of ad impressions and generating billions in revenue.

We are looking for a Software Development Engineer to join our team and help build an Advertiser Intelligence Center (AIC) service — the contextual backbone that powers our AI agents. You will design and implement the systems that aggregate, structure, and serve advertiser context to our agents, enabling AI-driven workflows that help account teams deliver better outcomes for advertisers at scale.This is a greenfield opportunity to define how advertiser knowledge is represented, retrieved, and reasoned over by autonomous agents. You will work at the intersection of large-scale data systems, Generative AI, and production agent frameworks to build context services that are fast, reliable, and rich enough to power intelligent sales workflows.

Why You Will Love This Opportunity

• Impact at scale: Your work powers AI agents used by thousands of account team members serving Amazon's largest advertisers globally.
• Greenfield architecture: The advertiser context center is being built now — you'll shape the data models, retrieval patterns, and APIs from the ground up.
• AI-native development: Work hands-on with agent frameworks, retrieval-augmented generation pipelines, and LLM-powered systems in production.
• Entrepreneurial team:We move fast, experiment often, and ship real products. Small team, big mandate.
• Career growth: Amazon Advertising is one of the fastest growing businesses at Amazon, with high visibility to senior leadership.

Key job responsibilities

• Design and build scalable services that aggregate, transform, and serve advertiser context (account history, campaign performance, deal data, behavioral signals) to AI agents in real time.
• Develop APIs and data retrieval layers that enable agents to access structured and unstructured advertiser intelligence with low latency and high reliability.
• Build and optimize retrieval-augmented generation pipelines that surface relevant advertiser context to LLM-based agents at inference time.
• Partner with Applied Scientists to integrate machine learning models (recommendations, segmentation, forecasting) into the agent context layer.
• Contribute to our agent architecture — including memory management, session isolation, tool orchestration, and guardrails for production agent deployments.
• Own end-to-end delivery of features from design through production, including operational excellence (monitoring, alarming, runbooks).