At Dataquest, I architected a Medallion Lakehouse on Databricks, migrating more than 600 TB of fragmented legacy storage into Delta Lake and Apache Iceberg. I also built automated compute spin-down and storage-tiering strategies that saved $140,000 annually.
I designed Airflow and Azure Data Factory pipelines for financial and transactional workloads, orchestrating 150+ daily DAGs with a 99.95% operational SLA. Query and Spark optimizations reduced cloud compute spend by 28%.
At Orderly Health, I built Kafka-based streaming pipelines processing over 20,000 clickstream messages per second with exactly-once semantics. I also developed Snowflake and Redshift transformation layers and reduced warehouse credit utilization by 35%.
Earlier, at Lendbuzz, I built data ingestion pipelines and migrated cron-managed Bash scripts and legacy SSIS packages to Apache Airflow, reducing pipeline failure points and operational overhead by 40%. My projects also include multi-cloud lakehouse integration and AI-ready data pipelines for RAG applications.
