Data Scientist

**Data Scientist - AI/ML Predictive Maintenance** **About the Role** We are seeking a Data Scientist to develop and deploy machine learning solutions for predictive maintenance, failure prediction and early-warning detection across critical infrastructure and industrial assets. The role focuses on analysing operational sensor data to identify equipment degradation, detect anomalies, forecast asset health and estimate time to failure. You will apply advanced machine learning techniques to solve real-world engineering challenges, particularly in environments where historical failure data is limited. The ideal candidate combines strong technical expertise in applied machine learning with a rigorous research mindset, sound experimentation skills and the ability to translate complex operational problems into practical, production-ready solutions. **Key Responsibilities** **1. Predictive Maintenance & Model Development** * Develop and validate machine learning models for equipment failure prediction, condition monitoring, early-warning detection, remaining useful life estimation and time-to-threshold forecasting. * Apply and evaluate techniques such as survival analysis, time-series forecasting, anomaly detection, change-point detection and representation learning. * Select modelling approaches based on measurable performance criteria, experimental evidence and operational requirements. * Balance detection accuracy, false-positive rates, model explainability and production deployment considerations. **2. Working with Limited Failure Data** * Develop modelling strategies for environments with limited or no historical failure events. * Address challenges involving imbalanced datasets, limited labelled failures, weak supervision, proxy labels and transfer learning. * Explore alternative data sources and approaches to improve model development where failure examples are scarce. * Establish robust validation methods to assess model effectiveness despite limited failure data. **3. Experimentation & Research** * Formulate hypotheses and design experiments to evaluate competing modelling approaches. * Establish baseline models, benchmark alternatives and measure improvements using objective performance metrics. * Document experimental results, assumptions, limitations and recommendations. * Justify model selection through data-driven evidence and reproducible experimentation. **4. Model Deployment & Monitoring** * Deploy machine learning models into development and production environments. * Monitor model performance, including precision, recall, data quality and model drift. * Define criteria for model retraining, rollback and retirement. * Investigate model performance degradation and implement corrective measures to maintain reliability. **5. Cross-Functional Collaboration** * Work closely with product managers, software engineers, platform engineers and engineering subject matter experts.