Applied Scientist, Alexa AI
Pubblicato il 22-08-2026 - Amazon in Torino
Description
As an Applied Scientist in the Alexa AI team, you will spearhead the advancement and deployment of state-of-the-art ML/RAG systems that revolutionize how millions of customers interact with Alexa. You'll leverage your expertise in machine learning, natural language processing, and large language models to create reliable, scalable, high-performance products that set new standards in operational excellence.
Working at the intersection of research and production, you'll translate latest AI innovations into customer-facing features that delight users daily. Your work will span the full ML lifecycle: from analyzing customer behavior patterns and building novel metrics for personal digital assistants, to deploying automated training pipelines and conducting rigorous A/B testing across diverse devices and endpoints.
Collaborating closely with business, engineering, and science teams across Amazon, you'll lead high-visibility programs that automate workflows and deliver measurable customer impact.
This role offers the unique opportunity to publish at top-tier conferences while seeing your innovations scale to one of the world's most popular voice assistants, serving millions of customers globally.
Key job responsibilities
As An Applied Scientist In The Alexa AI Team
You’ll analyze and model customer behavior at scale, building novel metrics for personal digital assistants across diverse devices and endpoints. Your work will involve creating deep learning, policy-based learning, and machine learning algorithms that directly impact customer experiences, translating complex data patterns into actionable insights that drive product innovation.
Your technical leadership will extend to building and deploying automated model training and evaluation pipelines, implementing complex machine learning and deep learning algorithms, and conducting rigorous model and data analysis through online A/B testing. You'll research and implement novel approaches that push the