Join us and contribute to driving excellence at MOTOLITE!
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.
- 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.
- 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.
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.
- 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.
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"