Find your next role

Description

Amazon is seeking a passionate and talented Applied Scientist to join the AERO team within Amazon PGX (Partner Growth and Experiences). Our mission is to elevate the experience of Selling Partners and Retail users through intelligent, collaborative Agentic AI solutions powered by specialized agents that deliver seamless, personalized support at scale.

As part of the AERO team, you will work alongside internationally recognized experts applying and advancing machine learning techniques to build Agentic AI solutions. Your work will directly impact millions of users in the form of products and services that improve their experience across Amazon's selling ecosystem.

Key job responsibilities
- Apply machine learning and statistical modeling techniques to build and improve Agentic AI components for Selling Partners and Retail users.
- Collaborate with senior Applied Scientists, Software Development Engineers, and Product Managers to design, prototype, and evaluate AI/ML models that power AERO's agentic workflows.
- Gain hands-on experience with Amazon's native and partnered Large Language Models (LLMs) and technologies such as AgentCore, AgentMemory, Strands, and Model Context Protocol (MCPs).
- Conduct experiments, analyze results, and iterate on model performance to improve agent task completion, accuracy, and business impact.

About the team
You will work closely with the AERO team to apply ML and LLM-based techniques to real-world vendor management and retail challenges. This includes running experiments on agent behavior, fine-tuning or prompting LLMs for specific use cases, evaluating agent outputs against business KPIs, and collaborating with engineers to integrate models into production agentic pipelines.