Lead Data Scientist with Prophet, Python

EPAM Systems; EE. UU.; A distancia; USD 140.000 - 190.000; Jornada completa

Descripción de la vacante

EPAM Systems is seeking an experienced data scientist focused on econometric and time-series modeling in the United States. You will build should-cost models, incorporate macroeconomic drivers, and relate commodity prices to spend data to provide actionable procurement insights.

You will ensure explainable outputs for stakeholders, employ rigorous cross-validation, and guard against data leakage and overfitting. A strong command of English is required.

Formación

• 5+ years in data science with econometric and time-series modeling.
• Skilled in Python, Prophet, and Chronos.
• Experience in procurement analytics including spend data, commodity pricing indexing.

Responsabilidades

• Build "should-cost" models through time-series and econometric methods
• Integrate macroeconomic factors and economic indicators as cost drivers
• Model relationships linking commodity prices, spend data, and external economic variables
• Ensure model outputs are interpretable to stakeholders by highlighting the factors behind each cost estimate
• Use time-series cross-validation techniques to guarantee model reliability
• Guard against data leakage and overfitting within time-series models
• Differentiates between correlation and cointegration, identifying when each is applicable
• Choose appropriate features for econometric and time-series models
• Utilize spend data and commodity price indexing to enhance procurement analytics

Conocimientos

Python

English proficiency

Herramientas

Prophet

Chronos

Descripción del empleo

Responsibilities

• Build "should-cost" models through time-series and econometric methods
• Integrate macroeconomic factors and economic indicators as cost drivers
• Model relationships linking commodity prices, spend data, and external economic variables
• Ensure model outputs are interpretable to stakeholders by highlighting the factors behind each cost estimate
• Use time-series cross-validation techniques to guarantee model reliability
• Guard against data leakage and overfitting within time-series models
• Differentiates between correlation and cointegration, identifying when each is applicable
• Choose appropriate features for econometric and time-series models
• Utilize spend data and commodity price indexing to enhance procurement analytics

Requirements

• 5+ years in data science, specializing in econometric and time-series modeling
• Skilled in Python, Prophet, and Chronos
• Background in procurement analytics, including spend data and commodity price indexing
• Knowledge of economic indicators, spend analysis, commodity pricing, and supplier diversity metrics
• Strong command of time-series cross-validation and explainable AI
• Demonstrated skill in preventing data leakage and overfitting in time-series models
• Clear grasp of the distinction between correlation and cointegration, and their respective use cases
• B2-level English proficiency or above