Senior Data Scientist, Predictive Maintenance (Industrial)

ONE SEVEN CONSULTING PTE. LTD.; Singapore; On-site; SGD 120,000 - 170,000; Full time

Job summary

ONE SEVEN CONSULTING PTE. LTD. seeks a Senior Data Scientist, Predictive Maintenance to build and run ML models that predict failures and warn of degradation in data centre assets.

You will use sensor data to forecast asset health and time to failure, and work across the tech stack to productionize solutions. You will benchmark a range of techniques, including anomaly detection and time-series forecasting, and design validation methods when failure events are scarce.

Qualifications

• Bachelor's or Master's degree in Engineering, Computer Science, Data Science or related field, AI/ML preferred.
• 5–8 years of relevant data science or applied ML experience in software or industrial settings.
• 2–3 years building condition monitoring, predictive maintenance or anomaly detection models on sensor data.
• Strong understanding of ML approaches, including time-series and anomaly detection, with rigorous validation.

Responsibilities

• Build and validate models on sensor data to predict failures and degradation.
• Benchmark techniques such as anomaly detection, survival analysis and time-series forecasting.
• Balance detection performance with explainability and production deployment needs.
• Develop validation methods for scarce failure events and manage model retraining criteria.
• Collaborate with product managers, software/ platform engineers and SME peers.

Skills

AI/ML fundamentals

Model evaluation

Experimentation

Production deployment

Education

Bachelor's or Master's in Engineering, CS or Data Science

Job description

ONE SEVEN CONSULTING PTE. LTD. seeks a Senior Data Scientist, Predictive Maintenance to build and run ML models that predict failures and warn of degradation in data centre assets.

You will use sensor data to forecast asset health and time to failure, and work across the tech stack to productionize solutions. You will benchmark a range of techniques, including anomaly detection and time-series forecasting, and design validation methods when failure events are scarce.