Research by Malathi Marineni Examines Agentic AI for Data Pipeline Failure Triage
Malathi Marineni
The research, “An Agentic AI System for Context-Grounded and Deterministic Failure Triage in Data Pipelines,” evaluates an approach designed to use relevant operational context when generating recommendations for pipeline failures. The work originated from an independent proof of concept focused on improving the consistency and speed of failure analysis.
According to the supplied research summary, the architecture was tested across 100 AWS Glue ETL failure scenarios. Key reported results include:
-96% top-1 recommendation accuracy
-97% top-3 recommendation hit rate
-Evaluation across 100 AWS Glue ETL failure scenarios
The study examines whether agentic AI can help technical teams identify likely causes of data pipeline failures while using controlled decision logic and available system context.
The approach is designed to support engineering investigation rather than replace technical judgment. Recommendations can be reviewed within existing operational processes before further action is taken.
The research also contributes to broader discussion around the use of AI in enterprise operations, particularly in environments where automated recommendations require defined controls and human review.
Marineni’s current professional work provides additional context for the research. Within Elevance Health’s Federal Employee Program business area, her work includes AI-assisted engineering and healthcare data modernization.
Recent areas of work include:
-Conversational AI analytics for business users
-AI-assisted extraction of legacy healthcare business rules
-AI-assisted software development practices within regulated environments
Her engineering background includes work on data platforms supporting large-scale healthcare operations. She has led the design and delivery of pipelines supporting the processing of approximately 360 million healthcare claims. She has also led the delivery of approximately 317 AWS and Snowflake data pipelines and related data assets.
In addition to the IEEE publication, Marineni has authored research addressing governance-aware large language model routing and agentic AI frameworks. She has also served as an IEEE peer reviewer.
Her academic background includes a Doctor of Pharmacy degree and a master’s degree in Computer and Information Systems Security. This interdisciplinary experience provides supporting context for her work across healthcare technology and applied artificial intelligence.
Further professional information is available through Malathi Marineni’s LinkedIn profile.
Malathi Marineni
Independent Researcher
email us here
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
