Research Scientist, CloudTune
Core
Develops sophisticated algorithms combining machine learning-based demand forecasting with mathematical optimization to solve large-scale capacity planning problems under uncertainty for AWS infrastructure.
Role type
Research Scientist (Forecasting & Optimization)
Builds
Automated procurement, retention, and release policies for Amazon's compute infrastructure
Domain
Cloud Infrastructure / Operations Research / Machine Learning
Deliverable
production ML models
Required skills
stochastic optimization, time series modeling, linear programming, mixed-integer programming, machine learning, data analysis, SQL
Preferred skills
forecasting techniques, data mining
Technologies
XGBoost, quantile regression, deep learning, R, SAS, Matlab
Responsibilities
Formulate capacity planning challenges as constrained optimization problems; Develop demand forecasting models with risk-aware loss functions; Design safety stock and retention policy optimization frameworks; Process and analyze large-scale operational data; Create technical documentation and present research findings
Seniority
Senior, hands-on IC