International Journal of Sustainability and Innovation in Engineering (IJSIE)
2025
DOI https://www.doi.org/10.56830/IJSIE202503
Authors
Abstract
In this paper, MLOps and data engineering are discussed with a focus on the role of ETLs throughout the ML pipeline.
With the growing adoption of AI solutions in organizations, the—to be used—integration of sound data management has become a crucial success factor influencing its effectiveness, reliability, and value.
The paper overviews the architectural strategies for integrating ETL into MLOps methodologies, introduces methods of automated feature engineering, and discusses main issues like data drift detection and versioning.
Through the analysis of the current trends and technologies, this paper outlines how integrated ETL is in the process of moving traditional ML projects from proving grounds to scalable production systems that are defined to deliver tangible business value.
Keywords: MLOps, Data Engineering, ETL Integration, Feature Pipeline Automation
Download Full Article 