data scientist for should-cost procurement models

The company builds data-driven “should-cost” models that estimate what goods and services ought to cost, using time-series and econometric methods to support procurement analytics.

• Develop “should-cost” models using time-series and econometric methods
• Incorporate macroeconomic factors and economic indicators as cost drivers
• Model relationships between commodity prices, spend data, and external economic variables
• Make model outputs explainable to stakeholders by showing which factors drive each cost estimate
• Apply time-series cross-validation techniques to ensure model robustness
• Prevent data leakage and overfitting in time-series models
• Distinguish between correlation and cointegration and determine when each applies
• Select relevant features for time-series and econometric models
• Work with spend data and commodity price indexing to support procurement analytics

• 3+ Years of experience in data science, focused on time-series and econometric modeling
• Proficiency in Python, Prophet, and Chronos
• Experience in procurement analytics, working with spend data and commodity price indexing
• Understanding of economic indicators, spend analysis, commodity pricing, and supplier diversity metrics
• Expertise in time-series cross-validation and explainable AI
• Proven ability to prevent data leakage and overfitting in time-series models
• Understanding of the difference between correlation and cointegration and when each applies
• English proficiency at B2 level or higher

• Условий в вакансии нет