Senior Lead Data Scientist Predictive AI & Time Series

Job Description

Senior Lead Data Scientist Predictive AI & Time Series

Position Summary

We are seeking a highly skilled and experienced Lead Data Scientist Predictive AI & Time Series to design, develop, and operationalize advanced predictive models and forecasting solutions for complex business problems.

The successful candidate will possess strong expertise in statistical modeling, machine learning, predictive analytics, and time-series forecasting, with demonstrated experience delivering solutions from initial problem formulation through production deployment and measurable business impact.

This position requires a combination of advanced analytical capability, strong software engineering discipline, business understanding, and scientific rigor. The individual will be expected to independently solve complex data science problems, evaluate alternative methodologies, and establish scalable, reliable, and explainable predictive solutions.

Key Responsibilities

Predictive AI & Advanced Analytics

• Design, develop, validate, and deploy predictive models addressing complex business and operational problems.
• Apply advanced statistical and machine learning techniques to predict future outcomes, behaviors, risks, demand, and business events.
• Develop solutions across regression, classification, propensity modeling, risk prediction, anomaly detection, and behavioral prediction.
• Identify meaningful predictive signals from large-scale structured, transactional, behavioral, and temporal datasets.
• Develop predictive models that support data-driven decision-making and measurable business outcomes.
• Evaluate model performance using both statistical measures and relevant business KPIs.
• Apply appropriate model explainability, interpretability, calibration, and uncertainty-estimation techniques.

Time Series Forecasting

• Lead the development of sophisticated forecasting solutions for univariate, multivariate, hierarchical, intermittent, sparse, seasonal, and non-stationary time-series data.
• Develop short-, medium-, and long-horizon forecasts based on business requirements.
• Analyze trend, seasonality, cyclicality, autocorrelation, structural breaks, external drivers, and temporal dependencies.
• Address forecasting challenges including missing observations, outliers, changing patterns, concept drift, forecast bias, and intermittent demand.
• Develop and evaluate statistical, machine learning, deep learning, and hybrid forecasting approaches.
• Incorporate relevant exogenous variables, including business, economic, operational, calendar, and event-driven signals.

Statistical & Machine Learning Modeling

• Select and apply appropriate modeling techniques based on problem characteristics, data availability, scalability, interpretability, and business requirements.
• Apply methods including, where appropriate:
  
• ARIMA / SARIMA / SARIMAX
• Exponential Smoothing / ETS
• State-Space Models
• Regression and Generalized Linear Models
• Random Forest
• Gradient Boosting
• XGBoost / LightGBM / CatBoost
• LSTM / GRU
• Transformer-based architectures
• Ensemble and hybrid approaches
• Probabilistic forecasting techniques
• Establish appropriate baseline models and demonstrate incremental improvement through rigorous experimentation.

Data Preparation & Feature Engineering

• Conduct exploratory and statistical analysis of complex datasets.
• Develop robust temporal and predictive features, including lag, rolling-window, trend, seasonality, calendar, event, and behavioral features.
• Identify and address data-quality issues, missing values, outliers, leakage, and temporal inconsistencies.
• Integrate internal and external data sources to improve predictive performance.
• Develop scalable approaches for feature generation and transformation.

Model Validation & Evaluation

• Establish scientifically rigorous model-development and validation methodologies.
• Implement time-aware cross-validation, rolling-window validation, walk-forward validation, and backtesting as appropriate.
• Evaluate forecasting models using metrics such as MAE, RMSE, MAPE, WAPE, sMAPE, MASE, and forecast bias.
• Evaluate predictive models using appropriate measures such as Precision, Recall, F1, ROC-AUC, PR-AUC, calibration, lift, and business-specific KPIs.
• Conduct residual, error, sensitivity, and segment-level analysis.
• Assess model robustness, stability, generalization, and performance under changing conditions.
• Ensure appropriate controls are implemented to prevent data and target leakage.

Productionization & MLOps

• Translate data science solutions into reliable, scalable production systems.
• Collaborate with Data Engineering, ML Engineering, Software Engineering, and Product teams to operationalize models.
• Develop production-ready training, inference, evaluation, and retraining workflows.
• Establish appropriate model monitoring and governance mechanisms.
• Monitor model accuracy, prediction quality, data quality, drift, forecast bias, and degradation.
• Support automated retraining and model lifecycle management.
• Ensure solutions are reproducible, maintainable, observable, and scalable.

Technical Leadership & Collaboration

• Independently lead complex data science initiatives from problem definition through implementation.
• Provide technical guidance on predictive modeling, forecasting methodologies, and analytical approaches.
• Collaborate with cross-functional stakeholders to translate business requirements into technically sound data science solutions.
• Clearly communicate complex analytical findings, assumptions, limitations, and recommendations to technical and non-technical audiences.
• Mentor Data Scientists and contribute to data science standards, methodologies, and best practices.
• Evaluate emerging technologies and research in Predictive AI, forecasting, machine learning, and deep learning.

Required Qualifications

• Bachelor's or Master's degree in Computer Science, Data Science, Statistics, Mathematics, Engineering, Operations Research, Economics, or a related quantitative discipline.
• 5+ years of professional experience in Data Science, Machine Learning, Predictive Analytics, or a closely related field.
• Demonstrated hands-on experience solving complex predictive modeling and/or time-series forecasting problems.
• Strong foundation in probability, statistics, statistical inference, optimization, and machine learning.
• Advanced proficiency in Python.
• Strong proficiency in SQL and experience working with large datasets.
• Strong experience with relevant Python data science and machine learning libraries.
• Demonstrated experience taking models from experimentation through production deployment.
• Strong analytical, problem-solving, and communication skills.

Preferred Qualifications

• Master's or PhD in a quantitative discipline.
• Demonstrated experience with large-scale, enterprise forecasting systems.
• Experience with hierarchical, probabilistic, or intermittent-demand forecasting.
• Experience with deep learning for sequential and temporal data.
• Experience with Transformer-based or foundation-model approaches for time series.
• Experience with causal inference or causal machine learning.
• Experience with PyTorch or TensorFlow.
• Experience with Spark / PySpark.
• Experience with AWS, Azure, GCP, or comparable cloud platforms.
• Experience with MLOps, model monitoring, experiment tracking, and automated model lifecycle management.
• Research publications, patents, open-source contributions, or demonstrated applied research experience are advantageous.

Technical Skills

Programming & Data: Python, SQL, Pandas, NumPy

Machine Learning: Scikit-learn, XGBoost, LightGBM, CatBoost

Statistical Modeling: Statsmodels, ARIMA, SARIMA, SARIMAX, ETS, State-Space Models

Deep Learning: PyTorch, TensorFlow, LSTM, GRU, Transformers

Forecasting: Time Series Analysis, Demand Forecasting, Multivariate Forecasting, Hierarchical Forecasting, Probabilistic Forecasting, Backtesting

MLOps & Engineering: MLflow, Model Monitoring, Model Deployment, CI/CD, Cloud Platforms, Production ML

Experience Expectations

Candidates should be able to demonstrate substantive ownership of real-world Predictive AI or Time Series initiatives.

During the selection process, candidates may be expected to explain:

1. The business problem and prediction objective.
2. The characteristics and limitations of the underlying data.
3. The statistical and machine learning approaches evaluated.
4. The rationale for model selection.
5. The feature-engineering strategy.
6. The validation and backtesting methodology.
7. The production architecture and deployment approach.
8. The model-monitoring and retraining strategy.
9. The limitations and trade-offs of the solution.
10. The measurable business impact achieved.

Experience limited primarily to academic assignments, basic forecasting exercises, Kaggle competitions, or consumption of pre-built forecasting libraries without substantive ownership will not be considered equivalent to production experience.

Core Competencies

• Advanced Data Science
• Predictive AI
• Time Series Forecasting
• Statistical Modeling
• Machine Learning
• Deep Learning
• Predictive Analytics
• Feature Engineering
• Experimental Design
• Model Validation
• Forecast Accuracy & Bias Analysis
• Production Machine Learning
• MLOps
• Analytical Problem Solving
• Business Acumen
• Technical Communication

Success in the Role

Success in this position will be demonstrated through the ability to:

• Solve ambiguous and technically complex predictive problems.
• Develop statistically rigorous and commercially relevant models.
• Improve forecasting and prediction performance through disciplined experimentation.
• Build solutions capable of operating reliably in production.
• Establish robust model monitoring and continuous-improvement practices.
• Translate advanced analytics into measurable business outcomes.
• Serve as a technical authority in Predictive AI, Machine Learning, and Time Series Forecasting.

Candidate Profile

We are looking for a hands-on Data Science professional with demonstrated production expertise, strong statistical foundations, and the ability to solve complex prediction and forecasting problems end-to-end.

The ideal candidate combines deep technical expertise, scientific rigor, engineering discipline, and business orientation to build predictive systems that are accurate, scalable, explainable, and impactful.