Hema Kommu
Data Engineer with 5+ years of experience specializing in Cloud-Native Data Architecture (Azure) for Real-time Analytics & ETL/ELT.
- Role
- Data Engineer at Citi
- Location
- Pittsburgh, PA, US
- LinkedIn followers
- 500 followers
About Hema Kommu
As a Data Engineer with over 5 years of experience, I specialize in building secure…
Experience
Data Engineer
May 2024 — Present
Developed scalable pipelines using Azure Data Factory (ADF) and Databricks to ingest, cleanse, and structure financial and customer behavioral data into Azure Data Lake-Engineered real-time data streaming using Kafka for transactional logs, reducing fraud detection time from hours to minutes across 20+ datasets-Implemented Delta Lake architecture (bronze/silver/gold) with ACID support, enabling controlled historical tracking and schema evolution-Created modular, parameter-driven ADF pipelines automating ingestion from SFTP, APIs, databases, and Blob storage, reducing manual intervention by 60% across 50+ workflows-Applied PySpark transformations for aggregated views and analytical KPIs for credit scoring and customer retention analysis-Designed Power BI dashboards for compliance and audit teams, visualizing suspicious activity and account anomalies-Integrated Azure Key Vault, RBAC, and Managed Identity for secure authentication and least-privilege access-Developed CI/CD pipelines in Azure DevOps, promoting reusable notebooks and ADF deployments via YAML and ARM templates-Tuned Spark cluster performance using dynamic partitioning, caching, and job parallelism, reducing heavy ETL job runtime by 40%-Implemented data quality checks and logging in ADF and Databricks to capture nulls, duplicates, and threshold alerts, pushing outputs to Log Analytics-Integrated REST APIs to ingest external credit bureau data, validating and joining scores into customer profiles-Led migration of legacy SSIS-based pipelines into Databricks jobs, ensuring fidelity and aligning with cloud modernization strategy-Collaborated on Azure Purview integration, defining classification tags, glossary terms, and lineage for regulatory datasets-Created job orchestration flows in Databricks with retry logic and dependency control, enhancing reliability-Developed metadata-driven frameworks for ingestion rules, file naming, and schema mapping using SQL tables.
Education
Robert Morris University
Master's degree
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.