Data Science & Machine Learning Manager

Role Overview

1. Process Understanding & Model Validation

• Develop a deep understanding of gas supply and processing workflows across upstream and midstream operations.
• Review and validate physical simulation models, ensuring alignment with operational and process constraints.
• Evaluate production forecasting models, including component-level outputs (C1, C2, C3, C4, C5+).

2. Feature Engineering Review

• Assess and validate feature engineering methodologies used in ML and deep learning pipelines.
• Ensure robustness, relevance, and completeness of input variables for predictive modeling.

3. AI/ML Solution Evaluation

• Review and validate AI-driven solutions leveraging blended Machine Learning and Deep Learning approaches.
• Translate complex technical outputs into clear, business-friendly insights for stakeholders.

4. Model Optimization & Performance Improvement

• Identify gaps in existing model designs and propose enhancements for improved accuracy and performance.
• Fine-tune model parameters, optimize feature importance, and prioritize constraints to achieve optimal outcomes.
• Recommend advanced optimization strategies for planning, scheduling, and commercial decision-making.

5. Commercial Impact & Operational Efficiency

• Challenge existing implementations to identify opportunities for improved efficiency and profitability.
• Drive data-driven strategies to optimize production outputs and business performance across the value chain.

Core Skills

• Advanced expertise in
• Python programming
• for data science and AI development.
• Strong experience in:

• Machine Learning (Regression, Classification, Clustering)

• Deep Learning (including sequence models, LSTM, attention mechanisms)
• Ensemble and hybrid modeling approaches
• Deep understanding of
• optimization techniques
• , including:

• Linear Programming (LP)

• Constraint-based optimization models
• Objective function optimization for profitability

Modeling & Analytical Expertise

• Hands-on experience with models such as:

• LSTM (Long Short-Term Memory)

• Kalman Filters
• SARIMA

• Support Vector Regression (SVR)

• Ridge Regression
• Probabilistic and statistical models
• Strong knowledge of:
• Model validation techniques
• Overfitting prevention and proper train-test segregation
• Hyperparameter tuning and performance optimization

Mathematical & Analytical Skills

• Strong ability to interpret and apply mathematical equations in modeling.
• Advanced problem-solving and analytical thinking capabilities.
• Expertise in evaluating hybrid AI architectures to derive optimal solutions.