Software Development Engineer, Unified Content Processing Pipelines

Deliver AI Innovation in Compliance at Amazon

Join a team building the future of automated compliance in the product space. Your work will directly impact Amazon's ability to serve customers and partners safely and efficiently across the globe.

We have a unique challenge to overcome : use ML and LLMs at the scale of Amazon. As a software development engineer you will develop scalable LLM infrastructure that can be used to process billions of documents and images each year, in a cost efficient way.

The ideal candidate will be eager to work in a team of innovators and contribute to a suite of tools that power automating and scalable content processing with the end goal of providing Amazon shoppers and sellers with a frictionless and trustworthy shopping and selling experience. You will be working in an Agile mode, collaborating with product management and partner teams to iterate at a fast pace and delight our end users.

Key job responsibilities

Your key responsibilities will include :

• 1) designing, developing and maintaining micro-services and Sage maker ML / LLM endpoints hosted in AWS;
• 2) adding new front-end features to existing tools;
• 3) improving performance and operational costs;
• 4) researching solutions for technically ambiguous problems;

5) mentoring other engineers.

About the team

Compliance Shared Services (CoSS) vision is to provide best-in-class automation of controls for internal compliance and business programs (our primary customers) to lower costs in the Amazon flywheel on behalf of our end customers : buyers and selling partners. We provide easy-to-use shared services with data privacy, security and tenant isolation across all stages, while continuously raising industry standards for automation performance and cost efficiency, giving our program customers confidence that they have access to optimal cost-efficient automation options to better serve their end customers.

BASIC QUALIFICATIONS

PREFERRED QUALIFICATIONS