Siva Sai Gavini
Senior Software Engineer | Multi-Cloud Data Engineer (AWS, Azure, GCP) | Spark/PySpark, Kafka, Airflow | Redshift/Snowflake/BigQuery/Scala | dbt | Terraform, Kubernetes, CI/CD | HIPAA, PCI-DSS, SOX
- Role
- Senior Software Engineer at Amgen
- Location
- Powell, OH, US
- LinkedIn followers
- 500 followers
About Siva Sai Gavini
I didn\'t plan to end up this deep in data pipelines. But here I am, and honestly, I…
Experience
Senior Software Engineer
Sep 2024 — Present · OH, US
I was brought in to make data trustworthy, fast, and compliant. I started by shadowing analysts and tracing lineage from source to dashboard to learn where batches failed and SLAs slipped. From that map I rebuilt the platform on AWS so curated data arrives predictably.I standardized landing and curated zones in Amazon S3, automated cataloging with AWS Glue, and exposed query paths through Amazon Athena and Amazon Redshift. For heavy transforms I used Apache Spark with PySpark, and I orchestrated ingest, checks, and publish stages with Apache Airflow. For near real time needs I introduced Apache Kafka so dashboards and models react in minutes. The result is clearer contracts, visible lineage, and alerts that catch issues early.Security and governance were designed in from day one. I partnered with InfoSec to apply least privilege IAM, encryption in transit and at rest using AWS KMS, and PHI access workflows aligned to HIPAA. I implemented data quality gates with freshness checks, schema and volume validations, and failure notifications that explain exactly what broke and why. I also trimmed cost with partitioning that matches business access patterns, compression, lifecycle rules for cold data, and tag based chargeback.To validate architectural choices I stood up a Snowflake and dbt pilot. I modeled high value transforms in dbt with source freshness tests and column level constraints, generated living documentation, and scheduled dbt from Airflow alongside existing pipelines. In parallel I compared TCO and performance for Snowflake and Redshift so each workload lands where it fits best.Everything ships through automation. I defined infrastructure as code with Terraform, wired CI and CD with GitHub Actions, and added run time observability with metrics and alerts. New environments come online in hours, analysts ask fewer is this table safe questions, and teams spend more time turning signals into decisions that move patient outcomes and operations forward now.
Education
University of Massachusetts Dartmouth
Master of Science - MS, Technology Management
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.