Director, Applied Science, Alexa for Shopping (Rufus)
Core
Leading the science vision and execution for Amazon's next-generation conversational AI platform (Alexa for Shopping/Rufus), focusing on multi-agent architectures powered by LLMs and SLMs to deliver personalized shopping experiences.
Role type
Director, Applied Science (Transformational Leadership)
Builds
Multi-agent AI systems for conversational commerce, integrating pre-purchase and post-purchase use cases across Amazon services.
Domain
E-commerce, Conversational AI, Large Language Models (LLMs), Reinforcement Learning (RL)
Deliverable
production ML models
Required skills
LLM-based architectures, post-training techniques (RLHF, DPO, fine-tuning), multi-agent systems, science strategy definition, team leadership of leaders, production AI/ML system scaling
Preferred skills
Ph.D. in quantitative field, thought leadership in LLMs/NLP/RL, deep technical judgment with business acumen, ability to influence senior executives
Technologies
LLMs, SLMs, Reinforcement Learning (RL), Fine-tuning, RLHF, DPO
Responsibilities
Define and execute science strategy for the conversational AI platform, lead a multidisciplinary organization of scientists and engineers, architect and scale multi-agent systems, partner with Product and Engineering to align AI investments, establish scientific best practices for experimentation and deployment, mentor and develop senior technical leaders
Seniority
Director, hands-on IC with heavy people management
