I built a production ingestion API at Jio Platforms for crash and profiling data, connecting ADB tooling, custom parsing, and database storage for daily QA test cycles. I also converted legacy raw-text outputs to structured JSON, reducing storage overhead by 30–40% and downstream parsing time by about 35%.
I've built end-to-end CDC, lakehouse, and real-time data pipelines using PostgreSQL, Debezium, Kafka, Iceberg, Databricks, Snowflake, dbt, and Airflow. My CDC Lakehouse Pipeline streamed database changes into Iceberg tables with 1.03s P95 latency, while my Databricks pipeline processed 355M+ taxi trip records and cut data scanned by 89%.
I enjoy designing incremental, analytics-ready data systems with reliable transformations, data quality checks, and reproducible deployments. I bring strong Python, SQL, cloud, and distributed-data foundations, along with a problem-solving record of 700+ LeetCode questions.
