•Built a distributed, end-to-end data pipeline execution framework with Ray, enabling customizable workflows and scaling ingestion and processing to terabyte-scale datasets (≈10× scale improvement).
•Developed and maintained backend services within a gRPC-based microservices architecture.
•Drove the platform’s migration to Parquet-based data storage, improving ingestion and downstream processing performance by nearly 4×.
•Worked extensively across AWS infrastructure to design reliable, cost-efficient backend systems.