Priyanka Deshpande

Data engineer with end-to-end experience building cloud-native ETL/ELT pipelines on Azure, delivering analytics platforms for large-scale datasets. Designed and delivered a scalable Azure-based data platform using ADF, Synapse, and ADLS Gen2, building end-to-end ETL/ELT pipelines with SQL and Python to enable reliable, high-performance data ingestion, transformation, and analytics for large-scale datasets. Developed data pipelines for ingestion, cleaning, transformation, and storage, creating structured datasets for analysis and reporting. Built and maintained dbt transformation pipelines across staging, intermediate, and mart layers — enforcing data quality with schema tests and source freshness checks to deliver reliable, analytics-ready datasets for compliance and reporting teams. Orchestrated end-to-end data pipelines using Apache Airflow DAGs, managing task dependencies, retry logic, and SLA alerts across ingestion, transformation, and model scoring workflows in production environments. Designed and optimized Snowflake data warehouses leveraging clustering keys, virtual warehouse separation, and external stages on cloud storage to reduce query latency and lower computing costs for analytics and ML workloads. Built machine learning models for classification problems using Scikit-learn and XGBoost, and evaluated performance using standard metrics such as Accuracy, Precision, Recall, and F1-score. Worked with unstructured data and applied LLM-based techniques to generate summaries and extract insights from documents. Created dashboards using Power BI and Tableau to present data trends and model results to stakeholders. Experienced in Azure cloud services and Agile/Scrum environments, supporting scalable data solutions and collaborative project delivery.