Join EY to build a remarkable career and contribute to a better working world. We are seeking a hands-on Lead Data Scientist – Manager to architect and deploy advanced analytical and machine learning solutions for the Oil & Gas industry. This role focuses on leveraging operational and engineering data to enhance performance diagnosis, outcome prediction, and decision-making across upstream and downstream assets.

Key responsibilities include developing sophisticated analytical solutions for well performance surveillance, production forecasting, and equipment health monitoring in upstream operations. In downstream, you will create inferential models for product quality and process optimization, analyzing data from process historians and lab systems. You will also be instrumental in defining engineering objectives, assessing data quality, engineering domain-informed features, and developing robust statistical and machine learning models.

Further duties involve establishing valid baselines and employing time-aware validation techniques to prevent data leakage. You will evaluate model performance using application-specific metrics, apply explainability and uncertainty quantification methods, and build reproducible training and inference pipelines. Documentation of assumptions, limitations, and retraining triggers, along with code reviews and mentoring junior data scientists, are also critical aspects of this role.

We require a degree in a relevant engineering discipline (petroleum, chemical, mechanical), statistics, mathematics, computer science, or data science. Demonstrated hands-on experience delivering Oil & Gas analytical solutions, with deep expertise in either upstream or downstream domains, is essential. A strong grasp of statistical principles, regression, classification, time-series analysis, feature engineering, model evaluation, and uncertainty estimation is necessary.

Candidates must possess experience handling challenges such as irregular sampling, missing measurements, limited failure labels, and evolving operating conditions. Evidence of personal contribution across the full data science lifecycle—from assessment and development to deployment and operational monitoring—is expected. Prior experience leading or reviewing the technical work of data scientists and collaborating effectively with engineering and technology teams is also a key qualification. Familiarity with Python, SQL, and core data science libraries like Pandas, NumPy, SciPy, and Scikit-learn is mandatory.