Predictive Analytics Engineer

Major Functions

• The Predictive Analytics Engineer is responsible for designing, building, and deploying predictive analytics solutions that improve asset reliability, enable predictive maintenance, and optimize operations and production across industrial facilities. The role centres on engineering robust, scalable, production-grade models and data pipelines, covering anomaly detection, failure prediction, remaining useful life estimation, and process optimization, that deliver measurable business value in complex Oil and Gas environments.
• The position requires strong domain expertise in Oil & Gas or Manufacturing, with a solid understanding of operational processes, equipment behaviour, asset performance management, and industrial data sources such as process historians, sensor/IoT data, maintenance management systems, and inspection records. The engineer will work closely with reliability, process, and maintenance engineers, subject matter experts, and business stakeholders to translate operational problems into analytical use cases, then validate, interpret, and embed the resulting solutions into day-to-day operations and decision-making. The role calls for both independent ownership of projects from concept to deployment and effective collaboration across multidisciplinary teams.

Essential Functions

• Design, build, and deploy predictive analytics solutions for industrial use cases including reliability monitoring, condition-based and predictive maintenance, failure prediction, and remaining useful life estimation.
• Develop and tune supervised and unsupervised models, such as regression, classification, clustering, anomaly detection, and time-series forecasting, using high-frequency sensor and operational data.
• Apply strong mathematical and statistical foundations (linear algebra, calculus, probability, statistics, signal processing) to feature engineering, model development, and optimization.
• Engineer scalable data and model pipelines using distributed and streaming technologies (e.g., Spark, Kafka, MapReduce) to process large volumes of process historian, IoT, and maintenance data
• Use Oil & Gas or Manufacturing domain knowledge to build physics-aware, context-driven predictive and prescriptive models that reflect real equipment behaviour and operating conditions.
• Work with data engineers, reliability and process engineers, and subject matter experts to identify, validate, cleanse, and interpret data sources such as historians, CMMS, inspection, and production records.
• Translate operational, maintenance, and production challenges into well-defined analytical use cases with clear success criteria and business value.
• Own predictive analytics initiatives end to end, from problem framing and data exploration through deployment, adoption, and ongoing performance monitoring.
• Develop rapid prototypes and proofs of concept in Python, R, or JavaScript. Experience with Java or Scala is an advantage.
• Design and implement MLOps practices, including CI/CD pipelines for models, automated testing, model versioning, containerisation, deployment automation, production monitoring, and model and data drift management.
• Ensure solutions are production-ready, reliable, explainable, and aligned with operational, safety, and integration constraints, including integration with existing dashboards, alerting, and asset performance management platforms.
• Communicate findings, model outputs, and recommendations clearly to technical and non-technical audiences, enabling operations and maintenance teams to act on predictive insights.

Minimum Educational & Professional Experience

• MS in Computer Science, Mechanical Engineering, Statistics, Engineering, or equivalent quantitative field
• Proven hands-on experience developing and deploying predictive models in an industrial setting, including regression, classification, anomaly detection, and time-series forecasting using supervised and unsupervised learning techniques.
• Strong mathematical and statistical foundation in linear algebra, calculus, probability, and statistics, applied to the analysis of sensor, process, and equipment performance data.
• Minimum 10 years of hands-on experience designing, developing, deploying, and operating predictive analytics and machine learning solutions for industrial use cases such as reliability monitoring, condition monitoring, and predictive maintenance.
• Mandatory experience in Oil & Gas or Manufacturing industry environments
• Demonstrated domain expertise in industrial operations, asset integrity and asset performance management, process optimization, or maintenance and reliability engineering.
• Proven track record of taking predictive models from prototype to sustained production use in real-world industrial settings, with measurable impact on uptime, maintenance cost, or production performance.
• Experience building scalable data and analytics pipelines using distributed and streaming technologies (e.g., Spark, Kafka, MapReduce) for high-volume sensor and time-series data.
• Experience working with industrial data sources such as process historians, SCADA/DCS, IoT platforms, and CMMS/EAM systems.
• Experience collaborating with cross-functional industrial stakeholders, including reliability, process, maintenance, and operations engineers and subject matter experts.
• Proficiency in Python, R, or JavaScript. Familiarity with Java or Scala is an advantage.
• Experience working in modern development environments, including version control, cloud platforms, and AI-assisted engineering workflows