Boca Raton, Florida, United States
Designed and led the delivery of a HIPAA-compliant healthcare lakehouse on Azure using ADLS Gen2, Azure Databricks, Azure Data Factory, and Azure Synapse Analytics, consolidating claims, eligibility, and EHR sources into a governed bronze/silver/gold medallion architecture. Built reusable ingestion frameworks for HL7/FHIR feeds, REST APIs, and flat files with Azure Data Factory, Azure Functions, Event Hubs, and Databricks Auto Loader to onboard new data sources through parameterized metadata-driven pipelines. Modeled core healthcare domains (members, providers, claims, encounters) into star and snowflake schemas in Synapse and Databricks SQL, implementing conformed dimensions and fact tables to support utilization, cost-of-care, and quality reporting. Developed PySpark and Spark SQL ELT jobs to cleanse, standardize, and de-duplicate member records, applying SCD Type 2 history, schema evolution, and rule-based data quality controls to produce trusted silver and gold Delta tables. Orchestrated end-to-end batch and near real-time workflows using Azure Data Factory pipelines and Databricks Workflows with SLA-aware scheduling, dependency management, retries, and alerts for daily and intraday loads. Enforced PHI/PII protection with Unity Catalog, Microsoft Purview lineage, Azure Key Vault, column-level masking, tokenization, and role-based access control aligned to HIPAA and GDPR requirements. Delivered interactive Power BI dashboards on curated gold datasets, providing business and clinical stakeholders with insights into claims cost drivers, risk scores, and provider performance. Established Azure Monitor and Log Analytics-based observability with automated alerting and integrated CI/CD via Azure DevOps, Git-based version control, and runbooks to support agile delivery and rapid incident remediation.
Built an omnichannel retail analytics platform on Google Cloud using Databricks on GCP, consolidating POS, catalog, transactional, and campaign feeds into a centralized lakehouse on Google Cloud Storage and BigQuery. Engineered batch and streaming ingestion with Google Pub/Sub, Dataproc, and Databricks Auto Loader to land data into a bronze layer on GCS with checkpointing and exactly-once guarantees. Designed retail star schemas and data marts in BigQuery and Databricks SQL for sales, basket, inventory, and campaign analytics, with conformed dimensions for store, product, customer, and time. Developed PySpark and Spark SQL pipelines to transform POS data into Delta tables, implementing SCD Type 2 logic, Delta Lake time travel, expectation-based data quality rules, and de-duplication for auditable, reliable datasets. Orchestrated the platform using Cloud Composer (Apache Airflow), Databricks Workflows, and Delta Live Tables to deliver declarative pipelines with built-in expectations, dependency management, and automated retries. Implemented near real-time catalog and pricing synchronization with Pub/Sub and Spark Structured Streaming, keeping product and pricing attributes aligned with store systems. Applied fine-grained governance with Unity Catalog and Google Dataplex, managing metadata, lineage, and IAM-based access for internal and vendor stakeholders. Integrated BigQuery and Snowflake datasets with Databricks pipelines to support A/B testing and exported curated experiment data to Python/Jupyter environments for statistical evaluation. . Built interactive sales and merchandising dashboards in Looker and Power BI over curated gold tables to visualize sales trends, basket composition, and campaign performance while improving reliability through retries, failover handling, Cloud Logging/Monitoring, Databricks job alerts, and GitHub Actions plus Terraform-based deployments.
Developed complex SQL queries and stored procedures across Oracle and SQL Server to integrate, cleanse, and standardize data for enterprise data marts supporting sales and marketing analytics. Created Power BI dashboards and Excel-based reports to track sales performance, campaign KPIs, and telecom marketing metrics, improving visibility for business stakeholders. Automated recurring Excel workflows using Python, reducing manual reporting effort and lowering the risk of data-entry errors. Supported Informatica PowerCenter ETL jobs by monitoring daily loads, performing incident triage, and refactoring SSIS packages to simplify maintenance and improve reliability. Implemented Python-based data validation checks for schema consistency, completeness, and business rules, increasing confidence in downstream reporting and analysis. Documented data dictionaries, lineage diagrams, and source-to-target mappings to meet compliance requirements and guide developers and analysts. Worked with QA teams in an Agile environment using JIRA for sprint tracking and defect management, translating stakeholder requirements into SQL logic and dashboard prototypes during sprint ceremonies. Built and maintained source-to-target mappings and incremental load logic in Informatica and SSIS to populate sales and marketing data marts, applying surrogate keys and slowly changing dimension patterns for historical tracking. Tuned SQL queries, indexes, and stored procedures on Oracle and SQL Server to reduce report refresh times, improving performance of high-volume sales and campaign reporting workloads. Performed data profiling and reconciliation between source systems and target marts to identify gaps, duplicates, and mismatches, ensuring accuracy of financial and marketing KPIs before publishing. Partnered with business analysts to gather reporting requirements, define metrics, and document data dictionaries, improving self-service adoption and reducing ad-hoc reporting requests.