Ajay Kamlapure

MEDICAID IMPLEMENTATION (FOR NEBRASKA, NEW MEXICO, WISCONSIN AND FLORIDA):
• Lead the design, development, and deployment of end-to-end data pipelines using Azure Data Lake and Azure Databricks to
support healthcare HEDIS Quality.
• Implemented scalable data warehousing solutions to handle large volumes of healthcare data for historical claims, encounters,
labs, pharmacy data.
• Collaborate with business and IT stakeholders to define data requirements for HEDIS reporting and other quality measures,
ensuring accurate and timely delivery within business defined timelines.
• Worked with Business for creating mapping documents for new historical pharmacy and claims dataset.
• Used Python and Pyspark to enhance the existing File Extraction model to generate the historical extract files in the HEDIS
required format.
• Troubleshoot and fix performance issues in Databricks jobs by reviewing the joins applying optimization techniques like
caching and partitioning delta tables.
• Used JSON rules for File name conventions and fixed issues while extracting the data from tables to text files.
• Drive data governance practices to ensure data quality, consistency, and compliance with HIPAA regulations.
• Mentored and lead a team of data engineers, providing technical guidance and ensuring best practices in data engineering
and project execution.
• Modified JSON to include new data sources in the existing data pipeline.
• Prepared Production deployment checklist and Production deployment instruction for smooth deployment of datasets for data
load jobs, extraction jobs and SFTP jobs.
• Collaborated with other teams for creation of Autosys scheduling jobs.
• Use of Autosys scheduling tool to trigger and monitor the workflows deployed in Databricks using POJO J2EE parameters.
• Use Azure Devops for code repository and follow the CI/CD deployment model for deploying into lower environments and work
with DLM team (Deployment Lifecycle Management) to deploy the code in Prod.
• Supported production issues. Flagged source data discrepancy in production by analyzing the dataset. And fixed the issues on
priority.