Data Scientist — Modelling

Join us and contribute to driving excellence at MOTOLITE!

Job Summary:

The Data Scientist — Modelling develops and validates statistical and machine-learning models that improve prediction, diagnosis, optimization and business decisions. The role is accountable not only for model quality but for whether the model changes a decision and produces measurable value.

3.  Key Duties and Responsibilities

Problem framing

• Frame modelling problems against a measurable business outcome and a specific operational decision point.
• Establish the baseline that the model must beat, and define what success and failure look like before development starts.
• Test new forecasting algorithms and machine-learning approaches against that baseline.

Model development

• Perform feature engineering, model development, validation, backtesting, explainability analysis and experiment design.
• Develop forecasting, quality, customer, maintenance, pricing and optimization models depending on assigned domain.
• Document assumptions, limitations, data requirements and failure modes.

Productionization and lifecycle

• Define model acceptance criteria, monitoring requirements, retraining triggers and retirement conditions.
• Work with data and AI engineers to integrate models into actual operational workflows rather than standalone notebooks.
• Monitor deployed model performance and drift, and act when performance degrades.

Value measurement

• Quantify incremental business value against the baseline after deployment.
• Report honestly where a model does not beat the baseline and recommend retirement.

4.  Key Deliverables

• Production models with documented model cards.
• Evaluation and backtesting reports.
• Monitoring plans, drift reports and retraining schedules.
• Business-impact measurement against baseline.
• Reproducible experiment records and feature documentation.

5.  Accountability and Success Measures

• Statistical validity and reproducibility of models.
• Explainability appropriate to the decision and the audience.
• Decision utility — whether the model changes what people do.
• Measured performance and value after deployment, not at development time.
• Honest reporting of model limitations.

6.  Working Relationships

• Internal: Data Engineers; AI / LLM Engineers; BI & AI Context Engineers; Data & AI Translators; domain and plant experts; Data Science & AI Capability Head.
• External: academic or vendor partners for specialized methods, as directed.

7.  Qualifications

Education

Bachelor's degree in Statistics, Mathematics, Computer Science, Engineering, Economics or a related quantitative field. Master's degree preferred for senior positions.

Experience

Three or more years developing and deploying statistical or machine-learning models with demonstrated production use. Manufacturing process, quality or demand-forecasting modelling experience preferred for site-assigned positions.

Certifications

Preferred: Databricks Machine Learning Associate or Professional. Optional: cloud machine-learning certification, specialized forecasting or reliability training.

8.  Technical Skills

• Python, including the scientific and machine-learning stack; SQL.
• Time-series forecasting; regression and classification; clustering and segmentation.
• Statistical process control and quality analytics for manufacturing-assigned positions.
• Modelling on image-derived and sensor-derived features produced by vision and edge services.
• Optimization methods and experiment design.
• MLflow or equivalent for tracking, registry and reproducibility.
• Model explainability, validation and drift monitoring.
• Databricks and distributed computation.

9.  Behavioural Competencies

• Scientific rigour and resistance to overfitting a narrative.
• Business framing — starts from the decision, not the algorithm.
• Clarity in explaining uncertainty to non-technical audiences.
• Willingness to kill a model that does not earn its keep.
• Collaboration with engineering to reach production.

10.  Level Guidance

Data Scientist

Develops models on defined problems with technical supervision. Three or more years of relevant experience.

Senior Data Scientist

Owns a modelling domain, sets validation standards, mentors and reviews. Six or more years of relevant experience.

Lead Data Scientist

Sets modelling methodology and lifecycle standards enterprise-wide and deputizes for the Capability Head. Nine or more years of relevant experience.

"Motolite offers you not just a job, but a career with boundless opportunities"