At Bank of Georgia, I architect and own production ETL pipelines for high-volume financial data using Apache Spark, Apache Airflow, and Python. I migrated core pipelines to an Iceberg/Polaris catalog architecture and moved orchestration from YARN to Kubernetes.
I also redesigned a core IFRS reporting table into a partitioned architecture and built incremental data pipelines. In a personal project, I built a GCP batch pipeline from Cloud Storage through Spark on Dataproc to BigQuery, orchestrated with Airflow and provisioned with Terraform.

