Senior Software Development Engineer, Ads Core Infra (ACI)
Amazon•Published 20 hours ago•First seen 4 hours ago
Advertisers will spend tens of billions of dollars this year leveraging Amazon Advertising to grow their business. We are looking for exceptional software engineers to build the next generation of intelligent data services that power AI-driven advertising experiences across the Amazon Advertising portfolio. As part of the advertising organization, our team focuses on delivering real-time data intelligence and advertiser context that enables AI agents and tools to make informed decisions on behalf of advertisers.
This work requires building redundant, highly available systems that scale to serve millions of advertisers across 20+ countries. Our services operate 24/7/365, providing sub-second access to advertiser intelligence that powers campaign recommendations, performance analysis, and automated optimization across all ad programs (Sponsored Ads and DSP).
The ARDS & Profiles team is responsible for two interconnected systems:
Ads AI Realtime Data Service (ARDS) is the analytical intelligence layer for all Amazon Advertising AI agents. We transform terabytes of advertising data (campaigns, ASINs, brands, budgets) into queryable functions that agents call in real-time to answer advertiser questions and drive automated decisions. We are building a near-realtime streaming (sub-1-minute data freshness), scaling from 14 to 20+ query functions, and expanding dataset coverage across Sponsored Ads and DSP programs simultaneously.
Ads Profiles is the personalization substrate for advertiser interactions. We compute and serve structured intelligence documents (advertiser profiles, brand profiles, campaign profiles, user profiles) that give AI agents deep context about who they are serving. This includes LLM-based inference for generating natural-language summaries, a federated contribution framework for partner teams to enrich profiles, and an MCP-based access layer for third-party agent consumption.
Our problem space covers:
Near-realtime data infrastructure: streaming pipelines (Kafka/Kinesis), incremental compaction, manifest-based query engines (DuckDB/Athena), and freshness monitoring with automatic fallback
Multi-tenant authorization: dataset-level permission resolution across multiple account types (Single Global Accounts, Manager Accounts) with configurable bypass mechanisms for internal agent consumers
LLM integration at scale: profile generation for 2.5M+ advertisers with hallucination detection, factual accuracy validation, and cost-optimized inference scheduling
Distributed systems: cross-region DynamoDB replication, regional failover, eventual consistency with strong read guarantees for authorization paths
Performance: P99 < 1 second at 40 TPS sustained for complex analytical queries; P99 < 100ms for profile serving via Fabric SDK
We stand up CI/CD pipelines with automated eval frameworks (300+ parameterized test cases, regression CI gates blocking deployment on correctness drops), integration testing via Hydra, and observability through per-table freshness probes and per-function latency instrumentation. Our engineers ship with confidence knowing that every query function and profile entity type has automated correctness validation before reaching production.
Our team uses AWS services including: DynamoDB, Lambda, ECS/Fargate, EMR (PySpark), Kinesis, S3, Athena, CloudWatch, CDK, CloudFormation, SQS, and Bedrock (Claude) for LLM inference. We build on internal Amazon infrastructure including Coral services, Apollo deployments, the Fabric SDK for agent consumption, and Minos for fine-grained authorization.
Key job responsibilities
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.
USA, NY, New York - 184,900.00 - 250,200.00 USD annually
This work requires building redundant, highly available systems that scale to serve millions of advertisers across 20+ countries. Our services operate 24/7/365, providing sub-second access to advertiser intelligence that powers campaign recommendations, performance analysis, and automated optimization across all ad programs (Sponsored Ads and DSP).
The ARDS & Profiles team is responsible for two interconnected systems:
Ads AI Realtime Data Service (ARDS) is the analytical intelligence layer for all Amazon Advertising AI agents. We transform terabytes of advertising data (campaigns, ASINs, brands, budgets) into queryable functions that agents call in real-time to answer advertiser questions and drive automated decisions. We are building a near-realtime streaming (sub-1-minute data freshness), scaling from 14 to 20+ query functions, and expanding dataset coverage across Sponsored Ads and DSP programs simultaneously.
Ads Profiles is the personalization substrate for advertiser interactions. We compute and serve structured intelligence documents (advertiser profiles, brand profiles, campaign profiles, user profiles) that give AI agents deep context about who they are serving. This includes LLM-based inference for generating natural-language summaries, a federated contribution framework for partner teams to enrich profiles, and an MCP-based access layer for third-party agent consumption.
Our problem space covers:
Near-realtime data infrastructure: streaming pipelines (Kafka/Kinesis), incremental compaction, manifest-based query engines (DuckDB/Athena), and freshness monitoring with automatic fallback
Multi-tenant authorization: dataset-level permission resolution across multiple account types (Single Global Accounts, Manager Accounts) with configurable bypass mechanisms for internal agent consumers
LLM integration at scale: profile generation for 2.5M+ advertisers with hallucination detection, factual accuracy validation, and cost-optimized inference scheduling
Distributed systems: cross-region DynamoDB replication, regional failover, eventual consistency with strong read guarantees for authorization paths
Performance: P99 < 1 second at 40 TPS sustained for complex analytical queries; P99 < 100ms for profile serving via Fabric SDK
We stand up CI/CD pipelines with automated eval frameworks (300+ parameterized test cases, regression CI gates blocking deployment on correctness drops), integration testing via Hydra, and observability through per-table freshness probes and per-function latency instrumentation. Our engineers ship with confidence knowing that every query function and profile entity type has automated correctness validation before reaching production.
Our team uses AWS services including: DynamoDB, Lambda, ECS/Fargate, EMR (PySpark), Kinesis, S3, Athena, CloudWatch, CDK, CloudFormation, SQS, and Bedrock (Claude) for LLM inference. We build on internal Amazon infrastructure including Coral services, Apollo deployments, the Fabric SDK for agent consumption, and Minos for fine-grained authorization.
Key job responsibilities
- Design, build, and operate real-time data retrieval services that power AI agent decision-making across Amazon Advertising, handling 100M+ API requests per day at sub-5-second latency
- Build and maintain streaming data pipelines that ingest petabyte-scale advertising and retail datasets, transforming raw signals into queryable intelligence functions with sub-10-minute data freshness
- Own the full lifecycle of advertiser profile generation, including LLM-powered entity resolution, knowledge graph construction, and real-time profile serving across 4+ entity types (advertisers, brands, ASINs, categories)
- Design and implement authorization and access control layers that enforce dataset-level permission granularity across multiple account types (Sponsored Ads, DSP, parent accounts) and advertising programs
- Develop evaluation frameworks that measure data quality, query accuracy, and agent decision effectiveness, using automated regression testing to maintain correctness as datasets and functions scale
- Build self-service onboarding infrastructure that allows partner teams (AI agents, campaign management, recommendations) to integrate new datasets and query functions without requiring core team engineering effort
- Operate production services at 99.99%+ availability across 20+ global marketplaces, owning on-call rotations, alarm tuning, runbooks, and incident response for tier-1 advertising infrastructure
- Drive technical design through written documents (design reviews, one-pagers, operational readiness reviews), collaborating with science teams on LLM integration and with data engineering teams on pipeline architecture
- Identify and eliminate scaling bottlenecks through load testing, profiling, and architectural optimization, targeting cost-efficiency improvements in compute, storage, and LLM inference spend
- Mentor engineers on the team through code reviews, design feedback, and architecture discussions, raising the technical bar across the organization
Basic Qualifications
- 5+ years of non-internship professional software development experience
- 5+ years of programming with at least one software programming language experience
- 5+ years of leading design or architecture (design patterns, reliability and scaling) of new and existing systems experience
- Experience as a mentor, tech lead or leading an engineering team
Preferred Qualifications
- 5+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience
- Bachelor's degree in computer science or equivalent
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.
The base salary range for this position is listed below. Your Amazon package will include sign-on payments and restricted stock units (RSUs). Final compensation will be determined based on factors including experience, qualifications, and location. Amazon also offers comprehensive benefits including health insurance (medical, dental, vision, prescription, Basic Life & AD&D insurance and option for Supplemental life plans, EAP, Mental Health Support, Medical Advice Line, Flexible Spending Accounts, Adoption and Surrogacy Reimbursement coverage), 401(k) matching, paid time off, and parental leave. Learn more about our benefits at https://amazon.jobs/en/benefits.
USA, NY, New York - 184,900.00 - 250,200.00 USD annually