SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 151,000 - 175,000 / annual
AssetWatch is seeking an experienced Machine Learning Engineer with specialized expertise in signal processing and predictive maintenance for industrial machinery. The role focuses on developing and deploying ML solutions that diagnose and predict faults in critical equipment such as bearings, gearboxes, shafts, motors, pumps, fans, and belts using data from IIoT sensors (accelerometers, temperature, electrical signals) and contextual machine data (images, metadata).
Key responsibilities include:
- Extract, preprocess, and analyze multi-modal sensor data to identify and enhance diagnostic features and fault-specific signatures from vibration, temperature, electrical, and image sources.
- Determine appropriate modeling techniques for each problem, including classical ML, supervised/unsupervised learning, anomaly detection, neural networks, deep learning, and signal-processing methods.
- Train, validate, and deploy predictive maintenance models using CNNs, RNNs, LSTMs, attention-based models, and time-series approaches.
- Develop and evaluate time-domain, frequency-domain, and contextual features, including harmonics, sidebands, envelope-spectrum analysis, and fault-specific indicators.
- Act as solution architect for ML initiatives: design end-to-end pipelines from data access through feature computation, training, inference, and production monitoring; document designs before large-scale execution.
- Build and scale distributed data and feature pipelines over large historical sensor datasets using Spark (EMR, Glue, SageMaker Processing), Athena, and Parquet-based layouts.
- Design and run data mining and annotation workflows connecting domain experts with data science, including case discovery, labeling, QA, and supervised training data generation.
- Build and maintain benchmark, training, validation, and regression datasets using real machine cases; establish ground truth and measurable model acceptance criteria with domain experts.
- Collaborate with MLOps and Data Engineering teams to deploy models on AWS using SageMaker (Training, Processing, Pipelines, Endpoints), S3, Athena, AWS Glue, EMR/Spark, Lambda, Step Functions, ECS/Fargate, Timestream, Aurora/RDS, Amazon Bedrock, and CloudWatch.
- Demonstrate strong software engineering fundamentals and data/ML engineering at scale with distributed processing.
- Communicate effectively with condition-monitoring domain experts, MLOps engineers, Data Engineers, Product, and other stakeholders.
Requirements:
- Thorough understanding of classical Machine Learning, Artificial Neural Networks, and Deep Learning algorithms (CNN, RNN, LSTM, attention mechanisms).
- Expertise in supervised and unsupervised learning, anomaly detection, ranking and decision models, feature engineering.
- Experience with Large Language Models (LLMs) and agentic AI.
- Hands-on proficiency with AWS services: SageMaker, S3, Athena, AWS Glue, EMR/Spark, Lambda, Step Functions, ECS/Fargate, Timestream, Aurora/RDS, Amazon Bedrock, CloudWatch.
- Strong software engineering fundamentals and experience with distributed processing (Spark or equivalent).
- Demonstrated experience developing and deploying ML solutions in AWS cloud environments.
- Ability to collaborate closely with MLOps and Data Engineering teams.
- Excellent communication and teamwork skills.
- Demonstrated research work in high-quality journals or significant applied work on related subjects is a plus.
