Agentic AI Scientist for Operational Orchestration

Amazon; Seattle, Northern (WA, KY); Hybrid; USD 143,000 - 193,000; Full time

Job summary

Amazon is seeking an Applied Scientist for Agentic AI within Amazon Fulfillment Technologies in Bellevue, WA. You will lead the development of agentic systems to assist with operational decision making and orchestration, and you will train LLMs using RL-based methods.

You will work on multi-agent orchestration, tool use, memory, and action execution, evaluate systems offline and online, mentor interns, and publish findings for external publication.

Qualifications

• PhD or Master's degree with 4+ years in CS/CE/ML or related field.
• 3+ years building models for business applications.
• Experience programming in Java, C++, Python or related language.
• Experience in algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing.
• Familiarity with Unix/Linux systems.
• Experience in professional software development.

Responsibilities

• Generating training and preference data for specific use cases (reasoning trajectories, tool traces)
• Reward modeling and policy optimization for LLMs: DPO, IPO, RLHF/RLAIF with PPO/GRPO, rejection sampling.
• Supervised fine-tuning on step-by-step trajectories and tool-use traces
• Verbal Reinforcement Learning and Continual Learning
• RL for LLMs, Offline RL and off-policy evaluation
• Agentic memory/state management; episodic and semantic memory; vector search; grounding with RAG.
• Evaluation: developing decision quality metrics, scaling LLM-based evaluations.

Skills

ML model development

RLHF/RLAIF

Distributed computing

Algorithms & data structures

Unix/Linux

Education

PhD or Master’s degree with 4+ years in CS/CE/ML

Tools

Java

C++

Python

Job description

Amazon is seeking an Applied Scientist for Agentic AI within Amazon Fulfillment Technologies in Bellevue, WA. You will lead the development of agentic systems to assist with operational decision making and orchestration, and you will train LLMs using RL-based methods.

You will work on multi-agent orchestration, tool use, memory, and action execution, evaluate systems offline and online, mentor interns, and publish findings for external publication.