Join Amazon's Finance Technology group to develop cutting-edge machine learning and agentic AI systems. You will analyze billions of dollars in daily transactions, directly impacting cost savings. Our team builds sophisticated detection systems to combat theft, fraud, abuse, and waste across Amazon's operations. We transform vast amounts of data into real-time risk decisions, scoring payments, identifying suspicious behavior, and uncovering complex fraud networks.

We are currently advancing our platform with generative and agentic AI, creating multi-agent systems that can discover fraud patterns independently from raw data. These systems collaborate with domain experts to implement production-ready detection logic, accelerating the deployment of new detection capabilities. This innovative approach allows new fraud defenses to be developed by both engineers and AI agents, guided by expert knowledge, and deployed in days rather than weeks.

Engage in the complete software development lifecycle, from initial design collaboration with customers and scientists to the construction of scalable, production-ready systems.

- Engineer ML pipelines capable of processing terabytes of data and scoring billions of dollars in transactions. - Develop agentic AI applications leveraging LLM-powered agents to identify risk patterns and generate detection logic. These agents will interact through defined tool interfaces with human oversight. - Construct the essential knowledge and memory systems for agents, including vector stores, knowledge graphs, and an entity-risk layer for collusion detection, alongside in-session and shared memory frameworks. - Define metrics for evaluating agent performance in fraud detection and transform analysts' ad-hoc investigation methods into reproducible, templated workflows. - Build a robust platform and tooling for governed data access, model deployment, and self-service rule authoring, enabling safe and rapid deployment of detection logic by both agents and non-engineers. - Guide models and agents from prototype stages to reliable, real-time production systems, incorporating direct customer feedback for continuous improvement.

This role requires a minimum of 3 years of professional software development experience, excluding internships. Additionally, you should have at least 2 years of experience in designing or architecting new and existing systems, focusing on design patterns, reliability, and scalability. Proficiency in at least one software programming language is essential.

Preferred qualifications include 3+ years of experience across the full software development lifecycle, encompassing coding standards, code reviews, source control management, build processes, testing, and operations. A Bachelor's degree in Computer Science or a related field is advantageous. Experience in machine learning, data mining, information retrieval, statistics, or natural language processing is highly beneficial. Familiarity with processing data using massively parallel technologies such as Redshift, Teradata, Netezza, Spark, or Hadoop-based big data solutions is also preferred.