Pavan Kumar Kattamuri

GCP FHIR store migration [CVS Health]

• Led the successful migration of a Google-managed encrypted FHIR store to a customer-managed encrypted FHIR store (CMEK-enabled) on Google Cloud Platform, supporting CVS Health’s regulatory alignment and data governance goals - HIPAA, HITECH, NIST SP 800-53 / 800-66, HITRUST
• 70 TB of longitudinal healthcare data, 45 FHIR resource types, covering a wide range of clinical, administrative, and patient data
• Zero-downtime migration strategy ensuring continuous availability for mission-critical workloads
• Seamless support for multiple upstream ingestion pipelines responsible for harmonizing and normalizing HL7v2, C-CDA, and other source formats
• Tech Stack: GCP Healthcare API, FHIR Store, CMEK, Cloud Dataflow, BigQuery, Pub/Sub, Cloud Spanner, IAM

Enterprise FHIR Platform [Highmark]

• Led multi-phase, ~$8M engagement with Google PSO and Highmark
• Ingested & harmonized 100+ TB of streaming HL7v2, FHIR, C-CDA, and CSV data
• Delivered longitudinal FHIR views via Healthcare API & BigQuery
• Integrated Redox feeds, reconciled data across 40+ FHIR resources
• Recognized by Highmark engineering leadership for agility & delivery

Teradata to BigQuery Migration [HCA Healthcare]

• Migrated 100+ TB, 100+ tables, and 100+ ELT jobs
• Enabled decommissioning of legacy Teradata stack, reducing costs
• Optimized data models for downstream BigQuery consumption

Azure Databricks to GCP Dataproc Migration [CVS Retail]

• Led migration of two critical retail analytics workloads (Promo FS & RxML) from Azure Databricks to GCP Dataproc, improving cost and runtime efficiency for CVS Health.
• Engineered optimized PySpark pipelines leveraging GCP-native orchestration, autoscaling, and ephemeral clusters, tailored for performance and cost-efficiency.
• Achieved up to 69% cost reduction and ~50% workload efficiency gain, delivering ~$50K in annual savings and reducing total job runtime by 6 hours.
• Success of this initiative catalyzed wider GCP adoption within CVS Retail, reinforcing Google Cloud’s strategic advantage for large-scale ML and forecasting use cases.

Realtime operating IoT DataLake [Adani]

• Designed and deployed real-time streaming pipelines using Cloud Dataflow to ingest and enrich sensor data from industrial IoT devices across Adani’s infrastructure.
• Delivered curated datasets to BigQuery and Looker, enabling real-time dashboards that improved operational visibility and reduced manual reporting efforts.
• Led cloud cost optimization efforts, architecting low-cost, high-performance GCP solutions that met SLAs while accelerating Adani’s GCP adoption.
• Established a scalable, cloud-native foundation for future IoT analytics, including knowledge graph exploration with Neo4j.

Imaging DataLake [Bayer]

• Built a centralized platform to ingest & de-identify DICOM imaging data to help radiologists view CT scans, MRI scans, X-rays, ultrasounds, and other modalities in near real-time
• Implemented PHI redaction using GCP Healthcare API
• Enabled fast access via Elasticsearch and BigQuery for radiologists

Hive to BigQuery Migration Toolkit [Google PSO]

• Built internal tool for Hive-to-BigQuery migration; published by Google on GitHub
• Designed incremental migration logic, fault recovery with Cloud SQL
• Co-authored internal best practices guide for PSO engineers
• Helped 10+ organizations to accelerate their Hive data warehouses to BigQuery with 60% reduction in migration time