dhakshanabalsan

Investigated production and UAT issues in enterprise applications, performing root-cause analysis to identify discrepancies and improve system reliability. Executed SQL-based data validation and reconciliation to maintain data integrity and accuracy across Guidewire PolicyCenter workflows. Monitored AI/ML model outputs and data pipelines to identify anomalies, performance degradation, and data quality issues. • Supported model retraining and optimization by validating input datasets, analyzing model outputs, and assisting with performance evaluation. • Assisted with AI/ML solution deployment, integration, and post-deployment validation across development, testing, and production environments. • Collaborated with technical teams and stakeholders to troubleshoot application and model-related issues and coordinate corrective actions. • Tracked defects and supported incident resolution using Jira within an Agile/Scrum environment. • Documented troubleshooting procedures, validation results, model performance observations, and resolution steps to support knowledge sharing and operational continuity

Developed and optimized backend data systems using PostgreSQL and Amazon S3, improving data retrieval efficiency and database performance. • Designed and implemented data pipelines and ETL workflows using PostgreSQL and Amazon S3, including data cleaning, transformation, and validation to ensure high data quality. • Built interactive dashboards using Apache Superset, enabling real-time insights and supporting data-driven decision-making. • Enhanced a machine learning-based recommendation system through feature engineering and model optimization, improving accuracy. • Optimized PostgreSQL queries, indexing, and schema design, reducing query execution time and improving system scalability. • Collaborated with cross-functional teams to integrate ML models and improve data workflows, utilizing Amazon S3 for scalable data storage and efficient data processing. • Investigated data pipeline failures and model input inconsistencies, performed root-cause analysis, and implemented corrective actions to improve data quality and workflow reliability. • Supported machine learning model evaluation and performance monitoring by analyzing outputs, validating datasets, and collaborating with the team on model improvements