Machine Learning Engineer - BC

Machine Learning Engineers help deliver machine learning solutions for industrial process environments: fault detection, predictive maintenance, quality optimization, and process control. You’ll work across the full project lifecycle: scoping problems with plant engineers, wrangling messy sensor data, building and deploying models, and making sure they work in production. 

Machine Learning Engineers demonstrate:

• High integrity
• A willingness to go beyond the ordinary to meet and exceed client expectations
• A desire for continual challenge and development, and excellent written and verbal communication skills

Reports To: Operations Director

JOB QUALIFICATIONS

Roles and responsibilities for this job may include, but are not limited to:

• Develop and deploy ML models (classification, regression, anomaly detection, time-series forecasting) for industrial process applications 
• Collaborate with process engineers and operators to translate domain problems into well-scoped ML tasks 
• Build robust data pipelines from historians, SCADA systems, and other industrial data sources 
• Design feature engineering strategies grounded in physical process understanding 
• Validate models against real plant conditions, not just offline metrics 
• Containerize and deploy models using Docker, with experience in Kubernetes or similar orchestration tools 
• Support model monitoring, retraining workflows, and CI/CD for ML pipelines
• Require domestic and international travel

Required Experience

• Degree in Engineering (Electrical, Mechanical, Chemical, or similar), Computer Science, or similar scientific/technical field

Ideal Experience

• 3-5 years of experience in applied ML or data science, ideally in manufacturing, process industries, or adjacent fields 
• Strong Python skills: scikit-learn, pandas, NumPy as a baseline 
• Experience with a range of ML approaches: gradient boosting (LightGBM, XGBoost), deep learning frameworks (PyTorch or TensorFlow), and unsupervised methods (clustering, autoencoders, anomaly detection) 
• ​​​​​​​Familiarity with time-series data and the challenges that come with it (irregular sampling, sensor drift, missing data, class imbalance) 
• ​​​​​​​Working understanding of process engineering fundamentals: heat/mass balance, process flow diagrams, and common unit operations 
• ​​​​​​​Practical experience with Docker; familiarity with Kubernetes, Helm, or cloud container services 
• ​​​​​​​Comfort working with messy, real-world data rather than clean benchmark datasets 
• ​​​​​​​Ability to communicate model results and limitations clearly to non-ML stakeholders 
• ​​​​​​​Must be eligible to work in the United States and Canada or able to obtain appropriate work authorization (visa sponsorship may be available) 
• ​​​​​​​Ability to travel domestically and internationally, including to industrial and manufacturing facilities 

• Experience with process control systems (DCS/PLC), control loop tuning, SCADA, and MES systems 
• ​​​​​​​Familiarity with OPC-UA, MQTT, PI Historian, or similar industrial data infrastructure 
• Exposure to Bayesian methods or probabilistic modeling 
• Experience with MLOps tooling (MLflow, Kubeflow, Airflow, or similar) 
• Experience deploying models in edge, on-premise, and cloud environments 
• Background in controls, chemical, mechanical, or process engineering