Machine Learning Engineer, Predictive Maintenance
About this role
Developing algorithms for the diagnosis, prognosis, anomaly detection, and health assessment of critical machinery components forms the core of daily work on this team. Engineers extract diagnostic features and fault signatures from IIoT sensors and other static data to build predictive maintenance solutions.
AssetWatch powers manufacturing uptime through an unparalleled condition monitoring experience, uniting engineers and business leaders to build the future of predictive maintenance. This role exists to design end-to-end machine learning solutions on AWS, handling everything from large-scale feature computation to production monitoring. A candidate with a specialized background in signal processing, predictive maintenance, and cloud architecture will thrive here.
What you will do
- Extract, preprocess, and analyze sensor data like machine vibration, temperature, and electrical signals to identify diagnostic fault signatures.
- Determine and apply appropriate modeling techniques including classical machine learning, deep learning, anomaly detection, and hybrid methods.
- Train, validate, and deploy predictive maintenance models using CNNs, RNNs, LSTMs, and attention-based architectures.
- Act as the solution architect for assigned initiatives by defining end-to-end design, execution strategies, and cost tradeoffs.
- Build and scale distributed data and feature pipelines over large historical sensor datasets using Spark, Athena, and Parquet layouts.
- Develop robust, modular, and maintainable machine learning software in Python alongside unit tests and version-controlled workflows.
- Collaborate closely with MLOps and Data Engineering teams to ensure smooth deployment, containerization, and monitoring of models in production.
What you need
- Hold a Master's or Ph.D. in Mechanical Engineering, Electrical Engineering, Computer Science/Engineering, or a related field.
- Demonstrate proven experience in predictive maintenance and condition monitoring focused on signal processing techniques such as FFT, Wavelet Transform, and Envelope Analysis.
- Possess strong foundations in machine learning algorithms, supervised and unsupervised learning, anomaly detection, and feature engineering.
- Show a strong understanding of time-series modeling and evaluation based on physical problems and operational requirements.
- Have hands-on experience architecting and delivering production machine learning solutions on AWS using services like SageMaker, S3, Athena, and EMR.
Skills and tech stack
Questions candidates ask
What is the salary range for this role?
The salary range for this position is $151,000 to $175,000 USD. Compensation is based on full-time employment.
What tech stack and cloud services will I use?
You will design end-to-end ML solutions on AWS using services such as SageMaker, S3, Athena, AWS Glue, EMR/Spark, Lambda, Step Functions, ECS/Fargate, Timestream, Aurora/RDS, Amazon Bedrock, and CloudWatch. Development involves Python and distributed data processing frameworks.
What kind of experience is required for this position?
Candidates need a Master's or Ph.D. in a related engineering or computer science field, along with proven experience in predictive maintenance, condition monitoring, and signal processing techniques. Hands-on experience architecting and delivering production ML solutions on AWS is also required.
What does the day-to-day work involve regarding model development?
The work involves extracting features from IIoT sensors to develop algorithms for diagnosis, prognosis, anomaly detection, and health assessment of machinery components. You will also act as a solution architect, defining end-to-end designs from data access to production monitoring.
The application is on the official AssetWatch, Inc. careers page.
Apply on company site