Are you interested in developing AI agents using state-of-the-art LLM techniques to revolutionize how Amazon optimizes its global inventory management? Join our team where we're applying the latest advancements in Generative AI to improve productivity and speed of decision making for Amazon Device Inventory Management!

The Amazon Demand Science Optimization organization is looking for an Applied Scientist with deep expertise in Machine Learning and Large Language Models to develop AI agents that provide data insights and automate the flow for inventory decision-making. Our team is responsible for science models that power world-wide inventory allocation for Amazon Devices business including Echo, Kindle, Fire Tablets, Amazon TVs, Fire TV sticks, Ring, and other smart home devices.

We're now leveraging the power of multi-agent systems where specialized AI agents collaborate to perform complex tasks – with agents dedicated to data analysis, insight generation, recommendation formulation, and decision explanation working in concert to provide unprecedented insights into our complex inventory management problems and make our models more explainable and effective.

Key job responsibilities
The successful candidate will be a self-starter, comfortable with ambiguity, with strong attention to detail, and an ability to work in a fast-paced and ever-changing environment and a desire to help shape the overall business.

Responsibilities include:
* Develop advanced AI agents using state-of-the-art techniques including:
- Retrieval-Augmented Generation (RAG) to enhance agent knowledge
- Multi-agent orchestration frameworks for complex problem-solving
- Chain-of-thought reasoning and reflection capabilities
- Tool use and tool learning for seamless interaction with optimization systems
- Planning and reasoning frameworks to handle complex multi-step tasks
- Agent memory and knowledge management across long-term operations
* Build LLM-powered systems that provide intuitive explanations of complex optimization models and decisions
* Create AI agents that can analyze large-scale inventory data, extract meaningful insights, and communicate them effectively
* Design and implement novel model explainability techniques using generative AI to make optimization models more transparent
* Establish scalable processes for agent benchmarking, validation, and implementation
* Collaborate with optimization scientists and engineers to enhance decision-making throughout the inventory management process

About the team
Amazon Science https://www.linkedin.com/showcase/amazonscience/posts/?feedView=all