VENKATESH NARISETTY

Client: GEHealthcare (GEHC)

Project: Enterprise Data Ingestion & Migration to GCP

• Role: GCP Data Engineer
• Summary:
• Led the design and implementation of scalable data pipelines to migrate on-prem Oracle data and ingest external API data into GCP.
• Key Contributions:
• Configured Fivetran connectors to extract Oracle data and stage it in Google Cloud Storage (GCS).
• Developed Python connectors for REST APIs with support for pagination, authentication, and retry logic.
• Built transformation pipelines using Dataflow and Dataproc (PySpark) to cleanse, enrich, and parse structured/semi-structured data.
• Integrated Cloud DLP API for masking/tokenizing sensitive data (e.g., PII) to meet compliance standards.
• Automated event-driven workflows using Cloud Functions triggered by file arrivals in GCS.
• Orchestrated ETL pipelines using Cloud Composer (Airflow) for reliable and scheduled execution.
• Loaded processed data into BigQuery raw layer; created logical views and stored procedures for business logic abstraction.
• Implemented CI/CD pipelines using Cloud Build, Git, and Shell scripts for automated deployment of Dataflow templates, Airflow DAGs, and infrastructure.
• Ensured data integrity, performance, and security across all ingestion and migration workflows.
• Technologies: Fivetran, REST APIs, GCS, Dataflow, Dataproc, BigQuery, Cloud Functions, Cloud Composer, DLP API, Cloud Build, GitHub, ServiceNow
• Languages: SQL, Python, PySpark, Shell Script, Apache Beam

Project: Real-Time Data Ingestion — GE Healthcare Manufacturing Unit

• Role: GCP Data Engineer
• Summary:
• Designed and implemented near real-time and real-time data pipelines to support manufacturing analytics.
• Near Real-Time Pipeline:
• Used Fivetran Webhook Connector to ingest data every 4 hours into BigQuery.
• Created BigQuery stored procedures to extract event names and populate live reporting tables.
• Automated transformations using Cloud Functions and orchestrated workflows with Cloud Composer.
• Real-Time Pipeline:
• Built a Cloud Run service to expose a webhook endpoint for ingesting live data.
• Integrated Cloud Pub/Sub for message handling and Dataflow (Apache Beam) for real-time processing.
• Achieved sub-minute latency for loading processed data into BigQuery, enabling real-time dashboards and alerts.
• Technologies: Cloud Run, Cloud Pub/Sub, Dataflow, Apache Beam, BigQuery, Cloud Functions, Cloud Composer, Stackdriver Logging, Error Reporting
• Languages: Python, SQL