Applied Scientist: Agentic AI & LLMs for Operations

Amazon Inc.; Cincinnati (OH); Hybrid; USD 143.000 - 193.000; Vollzeit

Benefits dieser Stelle

Health insurance

RSUs

401(k) matching

Paid time off

Zusammenfassung

Amazon Fulfillment Technologies (AFT) is seeking an Applied Scientist to lead the development of agentic AI systems for operational decision making and orchestration across Amazon buildings. You will work on multi-agent orchestration, tool use, memory, and action execution, and train LLMs using rejection sampling, SFT, continual post-training, and RL.

You will conduct rigorous offline and online evaluations, mentor interns, and publish findings to summarize progress for external publication.

Qualifikationen

• 3+ years of building models for business applications.
• PhD, or Master's plus 4+ years in CS/CE/ML or related field.
• Experience programming in Java, C++, Python or related language.
• Experience in algorithms, data structures, parsing, numerical optimization, data mining, parallel and distributed computing.

Aufgaben

• Generate training and preference data for reasoning trajectories and tool traces.
• Model reward and policy optimization for LLMs (DPO, IPO, RLHF/RLAIF).
• Fine-tune models on step-by-step trajectories and tool-use traces.
• Advance verbal reinforcement learning and continual learning.
• Apply RL for LLMs and offline RL with off-policy evaluation.
• Manage agentic memory, vector search, and grounding with RAG.
• Evaluate and scale LLM-based decisions.

Kenntnisse

Modeling experience

Programming languages

Algorithms & data structures

Experience with ML for business

Ausbildung

PhD

MS + 4+ years CS/ML

Tools

Unix/Linux

Jobbeschreibung

Amazon Fulfillment Technologies (AFT) is seeking an Applied Scientist to lead the development of agentic AI systems for operational decision making and orchestration across Amazon buildings. You will work on multi-agent orchestration, tool use, memory, and action execution, and train LLMs using rejection sampling, SFT, continual post-training, and RL.

You will conduct rigorous offline and online evaluations, mentor interns, and publish findings to summarize progress for external publication.