Asset Analytics Engineer

Capgemini is a global engineering services leader helping innovative companies with R&D and engineering across industries. The company combines AI, technology, and human ingenuity to transform operations, focusing on intelligent and sustainable futures. Capgemini works on digital and software technology solutions including autonomous cars and life-saving robots, employing over 410,000 team members globally.

• 5+ years of experience in data analytics, reliability engineering, or a related field with hands-on experience on platforms such as SmartSignal, Aspen Mtell, or AVEVA PRiSM, including Blueprint and model review workflows
• Strong proficiency in Python (especially Pandas, NumPy) for data manipulation and custom analytics development
• Solid SQL skills for querying CMMS systems (Maximo, SAP PM) and Historian databases
• Experience with industrial data systems including PI, OPC, and IP21 historians
• Familiarity with API data extraction tools such as Postman or equivalent
• Structured, detail-oriented approach to data quality and model governance with ability to manage model lifecycle activities including training, tuning, retraining, and performance monitoring across large asset fleets

• Perform complex tag mapping of historian data sources (PI, OPC, IP21) to the SmartSignal Standard Data Model, ensuring data lineage, quality, and consistency across fleet-level deployment
• Apply Similarity-Based Modelling (SBM) and Empirical Model Learning (EML) techniques to establish accurate "Normal" operating profiles and select Gold Standard training windows representing healthy asset states
• Develop and maintain Analytic Blueprints as reusable templates for common industrial asset classes (pumps, motors, transformers) to enable rapid and scalable deployment
• Monitor model performance using Precision/Recall metrics, perform retraining following asset overhauls, and tune statistical thresholds to minimize false positives and maximize signal fidelity
• Write Python scripts (using Pandas, NumPy) for data manipulation, custom analytic rule development, and feature engineering to enhance model accuracy