Drimva

Amazon

Applied Scientist II, Prime Video Personalization and Discovery Science, Prime Video - Personalization and Discovery Science

$143K–$223K

Sunnyvale, CA +2 · Full-time · Posted Aug 31, 2026

Location
Sunnyvale, CA
Seattle, WA
New York, NY
Experience
3+ years building models for business applications; PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience.
Education
PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience.

About the job

Develops machine-learning models and personalization solutions to help Prime Video customers discover relevant content, including titles with limited behavioral data.

What you'll do

  • Develop content-understanding and customer-behavior foundation models using deep learning, multimodal learning, and time-series techniques.
  • Design, implement, launch, and evaluate personalization solutions through offline and online experiments and data analysis.
  • Collaborate with engineers, product managers, and stakeholders; communicate findings and publish research.

What they ask for

  • 3+ years of building models for business application experience.
  • PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience.
  • Experience programming in Java, C++, Python or related language.
  • Experience in algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, or high-performance computing.

Nice to have

  • Experience using Unix/Linux.
  • Experience in professional software development.

Pay

USD 171,600.00 - 222,200.00 annually (base salary); USD 172,400.00 - 223,400.00 annually (base salary); USD 142,800.00 - 193,200.00 annually (base salary)

Tools

Java, C++, Python, Unix/Linux

AI focus

Build foundation models for video and text understanding, model customer behavior and viewing trajectories, and develop scalable recommendations and personalization, including for cold-start titles and niche interests.

Model-written summary · the original posting is authoritative

First seen Oct 15, 2026 · Last verified Nov 5, 2026 (UTC)