Accountabilities

• Collaborate closely with data scientists and engineering, manufacturing, and operations subject matter experts to design data pipelines that effectively support machine learning, deep learning, and advanced analytics use cases.
• Work with AVEVA PI AF, PI EF, and related PI tools to access, structure, transform, and prepare industrial data for reliable downstream analysis and ML-Ops applications.
• Ensure data engineering processes remain rigorous while supporting an agile approach to data validation, merging, transformation, and preparation based on the business use case being addressed.
• Define and monitor business-oriented data quality metrics, identify areas for optimization, and ensure that datasets remain reliable and fit for purpose.
• Analyze data cleanliness and identify potential dataset biases, gaps, or inconsistencies that could affect analytical or machine learning outcomes.
• Experiment with different data pipelines and processing approaches, building and optimizing solutions that extract maximum value from complex industrial datasets.
• Apply data mining techniques and state-of-the-art analytical methods to uncover meaningful patterns, correlations, and insights that can support data scientists and subject matter experts.
• Augment datasets through algorithmic methods or relevant third-party data sources when additional information is needed to strengthen engineering and manufacturing applications.
• Improve data collection procedures by identifying and incorporating information that can contribute to more effective automation, optimization, and operational decision-making systems.
• Work with data engineering specialists to ensure that data processing, cleansing, validation, and integrity checks are performed reliably and that ML-Ops data remains available for 24/7 operations.
• Present early data findings, mining results, and analytical insights clearly to technical and non-technical stakeholders.
• Collaborate with 24/7/365 ML-Ops teams to monitor the data feeding AI and machine learning models and help track data-related impacts on model performance over time.
• Integrate industrial time-series data with geospatial tools to improve data visibility and support more comprehensive industrial analysis.
•