Applied Scientist, Responsible AI for Alexa Safety

Amazon Inc.; Lombardia; Hybrid; EUR 110,000 - 170,000; Full time

Job summary

Amazon Inc. is seeking an Applied Scientist-II to advance Responsible AI for Alexa+. You will develop safety systems, train models that align with human values, and design evaluation methods to detect vulnerabilities in real-world conversations.

You’ll work with engineers and product teams to deploy trustworthy AI at scale across devices and languages. You will lead research that balances model capability with safety, building multi-agent architectures and automated testing to prevent issues

Qualifications

• Experience building machine learning models for business applications.
• Experience with patents or publications at top-tier conferences or journals.
• Proficient in Java, C++, Python or related languages.
• Experience with state-of-the-art deep learning model architectures, training, optimization, and pruning.

Responsibilities

• Develop Responsible AI solutions at scale for Alexa+ and related systems.
• Identify and mitigate AI safety risks early through multi-agent architectures and orchestration patterns.
• Collaborate with engineers and scientists to translate research into production-ready AI safety features.
• Mentor and contribute to a culture of responsible AI practices across teams.

Skills

ML model development

Patents/publications

Java/C++/Python

Deep learning architectures

Tools

Unix/Linux

Job description

Amazon Inc. is seeking an Applied Scientist-II to advance Responsible AI for Alexa+. You will develop safety systems, train models that align with human values, and design evaluation methods to detect vulnerabilities in real-world conversations.

You’ll work with engineers and product teams to deploy trustworthy AI at scale across devices and languages. You will lead research that balances model capability with safety, building multi-agent architectures and automated testing to prevent issues