Lead Applied Scientist: Personalization & Discovery

Amazon Science; Sunnyvale (CA); On-site; USD 192,000 - 260,000; Full time

Job summary

Amazon Science in Sunnyvale is seeking an Applied Scientist to own end-to-end machine learning models for Prime Video content understanding. You will work on foundation and time-series models, collaborate with cross-functional teams, and run rigorous offline and online experiments to drive product impact.

You will lead research and engineering efforts, publish findings, and stay current with advances in multimodal learning.

Qualifications

• 3+ years building ML models for business use
• PhD or Master+6+ years applied research required
• Proficiency in Java, C++, Python or equivalent languages
• Experience with neural deep learning methods and ML concepts

Responsibilities

• Develop foundation models for content understanding using deep learning and multimodal data
• Build time sequence models to predict user behavior and viewing trajectories
• Collaborate with engineers and PMs to deploy end-to-end solutions across Prime Video experiences
• Design offline and online experiments (A/B) and analyze results
• Communicate complex ideas clearly to teammates and stakeholders
• Publish research findings in top conferences/journals
• Contribute to continual improvement of modeling techniques and systems

Skills

Java

C++

Python

Deep learning

Multimodal learning

Education

PhD, or MS with 6+ years applied research

Tools

TensorFlow

Spark

Hadoop

Job description

Amazon Science in Sunnyvale is seeking an Applied Scientist to own end-to-end machine learning models for Prime Video content understanding. You will work on foundation and time-series models, collaborate with cross-functional teams, and run rigorous offline and online experiments to drive product impact.

You will lead research and engineering efforts, publish findings, and stay current with advances in multimodal learning.