Applied Machine Learning Developer

Product Performance Redlen’s Product Performance team is dedicated to ensuring our detector modules meet the highest standards of reliability, performance, and application suitability. The team plays a critical role in connecting detector design and manufacturing to real-world imaging outcomes by driving initiatives in reliability, image quality optimization, verification, validation & test (VV&T), and data-driven performance analysis. Leveraging advanced data science techniques, robust testing protocols, and cross-functional collaboration, Product Performance ensures that every detector we deliver supports world-class photon-counting CT imaging and exceeds customer expectations. Data Science The Data Science team builds the models, pipelines, and tools required to convert high-throughput detector data into actionable insight. By enabling scalable analysis of reliability, production, and in-field data, they support failure mode identification, performance characterization, and system optimization. They operate as a connective layer across Product Performance, Reliability, and Image Quality, ensuring decisions are grounded in quantitative evidence.   About the Role The Applied Machine Learning Developer develops machine learning, computer vision, and advanced analytical solutions for complex scientific, product-performance, reliability, sensor, and manufacturing data. The role works with multidisciplinary teams to translate technical problems into robust analytical solutions that improve defect detection, product characterization, process understanding, and data-driven decision-making. Responsibilities Machine Learning & Model Development Develop and validate machine learning models for classification, regression, clustering, anomaly detection, and prediction using complex scientific and manufacturing datasets. Design feature-engineering, model-selection, cross-validation, and evaluation strategies suitable for noisy, imbalanced, high-dimensional data. Evaluate when machine learning or deep learning provides meaningful value over established statistical, signal-processing, or algorithmic approaches. Computer Vision & Image Analysis Develop computer-vision and image-processing algorithms for automated detection, classification, segmentation, and characterization of spatial defects and patterns. Time-Series & Anomaly Detection Develop methods for time-series analysis, change detection, and anomaly detection to identify abnormal or degrading sensor behavior. ML Workflows & Software Development Develop reproducible and scalable machine learning workflows for data preparation, feature engineering, model training, validation, and deployment across local and cloud-based environments, including AWS. Translate successful experiments and notebooks into maintainable, reusable Python and MATLAB software and analytical tools. Model Validation & Statistical Analysis Quantify model uncertainty and false-positive/false-negative risk. Communicate model assumptions, limitations, and validation results clearly. Cross-Functional Collaboration Collaborate with data scientists, software developers, data engineers, scientists, technologists, and domain experts to define problems and deliver practical solutions. Other Other duties as assigned and relevant to the role.   Qualifications Required Experience Approximately 5+ years developing applied machine learning solutions. Analysis of time-series, spatial, sensor, imaging, or other high-dimensional scientific datasets. Computer vision and image-processing applications using OpenCV, scikit-image, PyTorch, TensorFlow, or comparable tools. SQL and structured analytical datasets within complex technical environments. Skills Strong proficiency in Python, including NumPy, pandas, SciPy, and scikit-learn. Strong understanding of supervised and unsupervised learning methods, including classification, regression, clustering, and anomaly detection. Solid statistical foundation, including cross-validation, precision and recall, class imbalance, uncertainty quantification, and data-leakage prevention. Strong software development practices, including version control, modular design, testing, code review, and technical documentation. Ability to translate ambiguous technical questions into well-defined analytical problems and independently develop appropriate analytical approaches. Familiarity with AWS services such as S3, Athena, Glue, SageMaker, Lambda, and ECR/ECS is considered an asset. Excellent verbal and written communication skills in English. Personal Attributes Strong analytical thinking and technical judgment. Comfortable working with ambiguity and independently advancing complex technical problems. Pragmatic approach to balancing scientific rigor with business and operational priorities. Preferred Experience Applied machine learning or advanced analytics supporting product performance, reliability, manufacturing, scientific, sensor, or other technically complex datasets. Scientific computing, algorithm development, signal processing, image processing, simulation, or experimental-data analysis using MATLAB. Cloud-based data processing and machine learning workflows in AWS environments. Modern data-lake and lakehouse architectures, including Parquet-based ecosystems. Deep learning techniques using PyTorch, TensorFlow, or equivalent frameworks. MLOps practices including Docker, MLflow, CI/CD, model deployment, and experiment tracking. Semiconductor, medical imaging, industrial sensing, scientific instrumentation, reliability, predictive maintenance, condition monitoring, or similar technical domains. Education Master’s degree in Computer Science, Data Science, Applied Mathematics, Statistics, Physics, Electrical Engineering, Computer Engineering, or another related quantitative field.   Salary Range: $100,000 - $125,000 CAD