Applied Scientist II, Search Ranking
Core
Invent universally applicable signals and algorithms for training machine-learned ranking models to improve search quality for Amazon's product search service.
Role type
Applied Scientist II, Search Ranking
Builds
High performance, fault-tolerant distributed search systems and production-level ranking components
Domain
E-commerce, Search, Machine Learning
Deliverable
production ML models
Required skills
algorithms and data structures, numerical optimization, data mining, parallel and distributed computing, high-performance computing, Java, C++, Python
Preferred skills
deep learning model architecture design, deep learning training and optimization, model pruning, MXNet, TensorFlow, Caffe, Pytorch, Unix/Linux, professional software development
Technologies
MXNet, TensorFlow, Caffe, Pytorch
Responsibilities
Analyze data and metrics from product search traffic, Design and deploy ML solutions for search ranking, Evaluate solutions via offline benchmarks and online A/B tests, Publish and present work at scientific venues
Seniority
Mid-level IC
