Data Analyst

I extract and integrate patient demographics, encounters, clinical records, and laboratory data from Epic Clarity into Amazon Redshift for population health analysis supporting more than 1.1 million cardiac patients. I design Redshift dimensional models for patient, encounter, diagnosis, and laboratory data, improving query performance by 32%.

I build high-risk cardiac cohorts using ICD-10 codes, Charlson Comorbidity Index criteria, and utilization indicators. I process clinical data with Python, Pandas, and NumPy; support predictive risk models with Scikit-learn; automate risk scoring on AWS EC2; and develop Tableau dashboards for risk stratification, readmission patterns, and care outreach. I also maintain HIPAA-aligned IAM access controls, Redshift security configurations, data lineage documentation, and audit support.

I built a procurement analytics warehouse on Azure Synapse, consolidating SAP MM, Oracle SCM Cloud, Microsoft Dynamics 365, Infor M3, JD Edwards, and legacy-system data supporting $2.8 billion in annual spend across 22 countries. I developed Azure Data Factory and Informatica pipelines to standardize more than 50 million monthly inventory records and improve data completeness.

I created demand forecasting models using Prophet, SARIMA, and XGBoost; supplier-performance and procurement-compliance dashboards in Power BI; inventory aging analytics using SQL and DAX; and automated savings reporting with Python, Pandas, and Power Automate. I implemented Power BI row-level security for more than 400 stakeholders and mentored junior analysts in SQL optimization, Power BI development, and Agile practices.

I developed a centralized loan portfolio reporting solution using SQL Server, integrating disbursement, repayment, delinquency, and collections data across four source systems for a loan portfolio exceeding ₹1,200 crore. I built and maintained 18 Power BI dashboards tracking portfolio health, PAR30/PAR90, NPA ratios, collection efficiency, and delinquency trends.

I automated daily disbursement and collection reconciliation using T-SQL and SSIS, created MIS reporting solutions with Excel Power Query and SQL Server, and developed a Python-based RFM customer segmentation model. I also performed cohort analysis, A/B testing, and statistical analysis to evaluate underwriting and collection strategies, while producing data dictionaries, ETL lineage documentation, and reporting specifications.

GPA: 3.5/4.0.