Cloud Capacity Forecasting & Optimization Scientist

Amazon; Seattle (WA); On-site; USD 126,000 - 170,000; Full time

Job summary

Amazon’s CloudTune team seeks a Research Scientist to advance forecasting and optimization for capacity planning of AWS infrastructure. You will formulate optimization problems, develop demand forecasts with ML approaches, and design policies to minimize costs while meeting service guarantees.

Collaborating with scientists, software engineers, data engineers, and product managers, you will work on stochastic optimization, time-series modeling, and operations research to scale compute

Qualifications

• PhD in a quantitative field or MS plus 1+ years of quantitative field research experience.
• Experience applying scientific principles to business problems and products.
• Experience with data analysis packages (R, SAS, Matlab, etc.).
• Experience using SQL databases to manage and analyze large data sets.

Responsibilities

• Formulate capacity planning and resource allocation challenges as mathematical optimization problems.
• Develop demand forecasting models using machine learning with risk-aware loss functions.
• Design optimization frameworks balancing holding costs and fulfillment risk.
• Process and analyze large-scale operational data to inform model development.
• Create, enhance, and maintain technical documentation and present findings.

Skills

Data analysis with R/SAS/Matlab

SQL data management

Forecasting models (ML)

Machine learning

Risk-aware loss functions

Education

PhD in quantitative field

MS + 1+ years research experience

Tools

XGBoost

SQL databases

Job description

Amazon’s CloudTune team seeks a Research Scientist to advance forecasting and optimization for capacity planning of AWS infrastructure. You will formulate optimization problems, develop demand forecasts with ML approaches, and design policies to minimize costs while meeting service guarantees.

Collaborating with scientists, software engineers, data engineers, and product managers, you will work on stochastic optimization, time-series modeling, and operations research to scale compute