Applied Scientist, AWS Agentic AI

About the job

The best coding assistants have to do two hard things together: understand code as the structured, interconnected system it really is, and keep learning from every interaction so they get better over time. Most AI tools today do neither well. They flatten code into tokens and lose the structure that makes software navigable, and they start each session from scratch, forgetting what they learned yesterday about a developer's codebase, patterns, and way of working. Amazon Web Services (AWS) is assembling an elite team of scientists and engineers to change both at once, and we are hiring an Applied Scientist who will work across code understanding and continual learning as a single, connected problem. You will help create generative AI technologies, including Kiro IDE and CLI, that are transforming how software gets built.

In this role you will turn the many signals around code, from static analysis and symbol graphs to build systems and language servers, into a unified representation our models can reason over, and you will make that system learn over time, so it accumulates knowledge without forgetting what it already knew, retrieves the right memory at the right moment, and personalizes the experience for each developer and team. These are not two separate jobs. A system that understands code deeply is also the one best positioned to remember and personalize how it helps, and you will own that intersection end to end. Your work will reach millions of developers rather than sit in a paper.