Data Scientist for AI/ML Predictive Maintenance Platform

Role Overview

We are seeking an experienced Data Scientist to develop and deploy machine learning tools aimed at failure prediction and early warning within data center and industrial asset environments. The focus centers on analyzing sensor data to identify degradation trends, anticipate asset health status, detect anomalies, and estimate time to failure on key equipment.

Key Responsibilities

• Create and validate predictive maintenance and early-warning models using operational sensor data for mechanical and electrical systems, including failure prediction and remaining useful life estimation.
• Employ and compare advanced techniques such as survival analysis, time-series forecasting, anomaly and change-point detection, plus representation learning methods.
• Use data-driven performance metrics and thorough experimental documentation to select optimal modeling approaches balancing accuracy, false positives, explainability, and production suitability.
• Develop solutions for scenarios with sparse or no failure examples, handling class imbalances, weak supervision, proxy labeling, transfer learning, and sourcing failure-data effectively.
• Design experiments to test multiple modeling hypotheses, establish baselines, evaluate alternatives, quantify improvements, and rigorously document results and constraints.
• Deploy models into development and live production systems, continuously monitoring key indicators like precision, recall, model drift, and data integrity.
• Define criteria and protocols for model retraining, rollback, retirement, and implement corrective measures upon performance degradation.
• Work closely with cross-functional teams including product management, software/platform engineers, and domain experts with mechanical and electrical engineering backgrounds to translate operational challenges into machine learning tasks and incorporate expert knowledge into models.

Candidate Requirements

• A Bachelor’s or Master’s degree in Engineering, Computer Science, Data Science, or closely related field; specialization in AI/ML strongly preferred. Knowledge or background in mechanical engineering or familiarity with industrial systems is beneficial.
• Between 4 to 7 years of practical experience in data science or applied machine learning within either large software firms or industrial engineering settings.
• At least 2 to 3 years of direct experience building predictive maintenance solutions including failure prediction, anomaly detection, and condition monitoring with engineering or IoT sensor datasets.
• Proficient understanding of standard machine learning techniques such as time-series modeling, anomaly detection methods, classification and regression algorithms, and sound practices in evaluating model performance.

Business Information

Business Segment: Connectivity

Platform: Operating Division