Data Scientist - Predictive Maintenance

Cutsforth•Ferndale, WA

4d•Remote

About The Position

Applies data science and machine learning to the analysis of electrical, vibration, and acoustic signals, transforming raw time-series sensor data into actionable diagnostics and predictive insights for rotating industrial equipment. Partners with engineering and domain experts to design and deploy production-grade signal processing and ML solutions for predictive maintenance across industrial applications. Operates effectively in ambiguous problem spaces where signal quality, environmental noise, and domain constraints require both technical rigor and adaptive thinking.

Requirements

Bachelor’s degree in Electrical Engineering, Computer Engineering, Physics, Applied Mathematics, Acoustical Engineering, Mechanical Engineering, Aerospace Engineering, or a closely related engineering discipline required.
5+ years of professional experience in data science, machine learning, or applied signal processing, with demonstrated work on electrical, current/voltage, or industrial sensor signal data.
Direct industry experience in one or more of: Industrial/Rotating Equipment, Power Systems, Electrical Machine Diagnostics, or Condition Monitoring.
Hands-on experience with time-series and signal processing techniques, including spectral analysis, filtering, and feature extraction from raw sensor data.
Proficiency in Python, including scientific computing libraries (NumPy, SciPy, pandas) and ML frameworks (scikit-learn, PyTorch, or TensorFlow).
Familiarity with electrical measurement and analysis workflows (e.g., current/voltage waveform capture, power quality analyzers, or equivalent instrumentation).
Strong analytical and problem-solving skills with the capacity to work through ambiguous or data-sparse problem spaces.
Excellent written and verbal communication skills; ability to present technical findings to non-technical audiences.
Successfully pass background check for cybersecurity site access.
Strong foundation in signal processing theory and application, including experience with electrical, acoustic, or time-series data in a professional setting.
Proficiency in Python for data manipulation, signal processing, and model development (NumPy, SciPy, pandas, scikit-learn, PyTorch or TensorFlow).
Ability to work with uncertainty and incomplete information; comfortable forming and testing hypotheses when ground truth is limited.
Clear communicator capable of translating technical signal processing and ML findings to non-specialist audiences.
Self-directed and effective working remotely across cross-functional teams.
Must reside in the United States; not accepting applicants in California, Illinois, or New York.

Nice To Haves

Master’s degree in Electrical Engineering, Computer Engineering, Physics, Applied Mathematics, Data Science, or a related field.
Experience with Electrical Signature Analysis (ESA), Motor Current Signature Analysis (MCSA), or similar electrical machine diagnostic techniques.
Familiarity with rotating machinery fault physics (bearing fault frequencies, eccentricity, winding faults, broken rotor bars).
Demonstrated ability to own an ML model from prototype through production, including monitoring and retraining.
Familiarity with array/multi-sensor signal fusion across electrical and vibration domains.
Familiarity with cloud platforms (AWS, Azure, GCP) and MLOps tooling (MLflow, Docker, Airflow, CI/CD pipelines).
Experience with physics-informed modeling approaches.
Active participation in the broader signal processing or data science community through publications, open-source projects, or conference presentations.

Responsibilities

Design and develop signal processing pipelines and machine learning models that operate on electrical (current/voltage), vibration, and acoustic time-series sensor data, including symmetrical component analysis, matched filtering, wavelet decomposition, and time-frequency analysis techniques.
Evaluate algorithm performance using both objective metrics and subjective measures, including integration with speech recognition engines where applicable.
Perform exploratory data analysis, feature engineering, and signal feature extraction on raw electrical, vibration, and acoustic data to surface fault patterns and anomalies.
Analyze and interpret signals from electrical asset monitoring systems (motors, generators, pumps) utilizing electrical signature analysis, vibration analysis, and signal processing expertise to support fault isolation and anomaly detection.
Use cross-sensor asset monitoring data (temperature, speed, load) to characterize and validate signal-derived diagnostics.
Apply data-driven signal processing methods to characterize and isolate faults at the subsystem, component, and machine level, identifying root causes from spectral, electrical, and vibration sensor data in rotating industrial equipment.
Contribute to end-to-end ML workflows including data ingestion, model training, inference, and monitoring for drift and degradation in live environments.
Collaborate with engineering, product, and domain SMEs to translate operational challenges into well-scoped data science solutions.
Communicate findings, model performance, and business value clearly through visualizations, written documentation, and presentations to technical and non-technical stakeholders.
Explore and evaluate emerging signal processing and AI techniques, recommending production incorporation where appropriate.
Candidate will be responsible for reviewing policies and procedures related to cybersecurity and those relevant to the functions of their role.
Candidate is expected to maintain a cybersecure work environment.