Senior ML Engineer, Predictive Maintenance for Wind Turbines

The challenge

A single main bearing replacement costs $350K and takes the turbine offline for two weeks. The fleet sees dozens of unplanned bearing failures per year. A model that catches degradation 90 days before failure -- instead of the current 30 -- changes the maintenance economics of the entire fleet.

Role overview

Machine learning models that predict equipment failures in wind turbines, trained on SCADA data, vibration sensor readings, and 8 years of maintenance records. The models don't live in a research notebook. They run in production, and operations dispatches crane crews based on their output. When the model is wrong, a $350K repair becomes a $600K emergency.

Specializations

Technical requirements

• MS or PhD in Machine Learning, Statistics, or Engineering with ML focus
• 5+ years applying ML to physical systems (manufacturing, energy, aerospace, or industrial)
• Experience with time-series anomaly detection and remaining useful life prediction
• Proficiency in Python (scikit-learn, PyTorch or TensorFlow, pandas)
• Experience deploying ML models to production with monitoring and retraining pipelines
• Ability to work with domain experts who think in terms of physics, not features

Key responsibilities

• Develop and deploy ML models for predicting bearing, gearbox, and generator failures in wind turbines
• Design feature engineering pipelines that extract predictive signals from 10-minute SCADA data and high-frequency vibration sensors
• Build anomaly detection systems that identify degradation patterns across thousands of turbines
• Validate model performance against historical failure data and track prediction accuracy over time
• Collaborate with reliability engineers to translate model outputs into actionable maintenance decisions
• Deploy models to production using cloud infrastructure (AWS SageMaker or equivalent)

Compensation & Benefits

• Base salary: $140,000 - $180,000
• Annual bonus target 10%
• Stock purchase plan at 15% discount
• Remote-first with optional Denver office access
• Conference attendance budget ($5,000/year)