OT Artificial Intelligence Machine Learning Engineer

We are seeking an OT Artificial Intelligence Machine Learning Engineer to design, develop, deploy and operate production-grade machine learning solutions for industrial environments. This role turns operational problems and industrial data into dependable models for computer vision, predictive and prescriptive analytics, anomaly detection, generative AI, agentic systems and physical AI.

This position is part of the OT Artificial Intelligence Team and provides hands-on machine learning engineering leadership and delivery support for OT AI use cases. The role builds reusable model-development, validation, packaging, deployment and monitoring patterns across edge, on-premises and cloud environments, with particular attention to safety, reliability, cybersecurity and operational supportability.

Our team is highly technical, creative and cross-functional. The Machine Learning Engineer will work with OT domain experts, data engineers, data scientists, solution architects, cybersecurity teams and operations personnel to move models from experimentation into sustained operational use. Success means delivering measurable AI outcomes while ensuring models behave predictably under real asset conditions, including limited connectivity, changing processes, constrained compute and imperfect data.

HIGH LEVEL JOB DESCRIPTION/SCOPE

This is a dedicated machine learning engineering role supporting OT AI. It covers model design, feature engineering, training, evaluation, optimization, deployment, monitoring, retraining and retirement for AI solutions used in industrial operations. The role supports edge and centralized inference across time-series, event, image, video, document and multimodal data while maintaining human oversight and safe fallback behavior.

Key Responsibilities

• Translate industrial business problems into machine learning objectives, measurable success criteria and technically viable solution designs.
• Develop, train and evaluate machine learning models for computer vision, anomaly detection, forecasting, predictive maintenance, optimization, natural language and multimodal use cases.
• Engineer features and model inputs from time-series data, events, alarms, sensor telemetry, images, video, documents and contextual asset data.
• Establish reproducible experimentation, model versioning, evaluation and approval workflows using MLOps practices.
• Package and deploy models for secure inference at the industrial edge, on-premises and in approved cloud environments.
• Optimize models for latency, accuracy, memory, compute, energy consumption and bandwidth constraints in operational environments.
• Design rigorous validation approaches using representative operating conditions, rare events, failure scenarios, degraded data and disconnected operation.
• Implement production monitoring for model performance, confidence, drift, bias, data quality, resource utilization and downstream operational impact.
• Define retraining triggers, rollback procedures, fallback behavior and model retirement criteria in partnership with architecture, cybersecurity and operations teams.
• Build explainability, traceability, human review and audit mechanisms appropriate to the operational risk of each AI solution.
• Partner with data engineers and domain experts to define training datasets, labeling standards, ground truth, sampling strategies and synthetic-data needs.
• Produce model cards, technical documentation, test evidence, deployment runbooks and operational support procedures.
• Contribute actively within an embedded Agile team and continuously improve shared OT AI engineering standards, components and practices.

Required

• BS in Computer Science, Machine Learning, Artificial Intelligence, Data Science, Engineering or a related field, or equivalent experience.
• 5 or more years of experience developing and deploying production machine learning or AI solutions.
• Strong programming skills in Python and experience with production software engineering, testing, source control, code review and CI/CD.
• Hands-on experience with machine learning frameworks such as PyTorch, TensorFlow, scikit-learn or equivalent technologies.
• Experience with feature engineering, model training, hyperparameter tuning, evaluation, error analysis and reproducible experimentation.
• Experience deploying and monitoring models in production using containers, APIs, orchestration, model registries and automated delivery pipelines.
• Experience detecting model and data drift, diagnosing performance degradation, and implementing retraining, rollback and retirement processes.
• Working knowledge of industrial data such as sensor telemetry, time-series data, alarms, events, images, video, maintenance records or engineering information.
• Understanding of OT cybersecurity, network segmentation, least-privilege access and the constraints of operational environments.
• Ability to communicate model behavior, uncertainty, limitations and operational risk to engineers, operators, architects, cybersecurity reviewers and business sponsors.
• Demonstrated ownership of production AI services, including troubleshooting, incident response, documentation and continuous improvement.

Preferred

• MS in Computer Science, Machine Learning, Artificial Intelligence, Data Science or a related field.
• Experience with Microsoft Azure AI services, Azure Machine Learning, Microsoft Fabric, Azure Databricks or equivalent platforms.
• Experience developing and deploying computer vision models for industrial inspection, safety, reliability or autonomous systems.
• Experience with forecasting, anomaly detection, predictive maintenance, optimization or prescriptive analytics for industrial assets and processes.
• Experience with generative AI, retrieval-augmented generation, agentic systems, multimodal models or small language models deployed in controlled environments.
• Experience deploying optimized inference at the edge using GPUs, NPUs or other AI accelerators in constrained or intermittently connected settings.
• Experience with model compression, quantization, pruning, distillation, ONNX or hardware-aware optimization.
• Experience with responsible AI, explainability, model risk management, bias testing, adversarial testing or AI assurance.
• Familiarity with industrial IoT platforms, historians, OPC UA, MQTT, digital twins, knowledge graphs or asset-context models.
• Upstream, midstream or downstream domain experience and comfort working directly with asset and operations personnel.

CRITICAL SELECTION CRITERIA

• Machine Learning Capability - Demonstrated ability to develop, validate, deploy and sustain production machine learning models using disciplined software engineering and MLOps practices.
• Operational Reliability - Ability to design models for industrial constraints, detect degraded behavior, and implement effective monitoring, human oversight, fallback and rollback mechanisms.
• OT and AI Understanding - Knowledge of industrial operations, OT data and cybersecurity requirements, and the full lifecycle needed to move models from experimentation into dependable operational use.
• Collaboration and Communication - Ability to translate operational problems into measurable ML objectives and clearly explain model performance, uncertainty, trade-offs and risks to technical and nontechnical stakeholders.