At RAMP, I built scalable ingestion pipelines that brought more than 50,000 unstructured documents into a centralized data lakehouse for an internal AI assistant. I also developed permission-aware embedding and vector indexing workflows for RAG retrieval.
I implemented a Bronze-Silver-Gold data architecture with Databricks Spark and Delta Lake, transforming documents into standardized chunks and reducing redundant re-embedding work by 39%. I automated pipeline orchestration and monitoring, achieving 99.5% pipeline reliability.
At TATA CONSULTANCY SERVICES, I worked on Best Buy data engineering and inventory analytics projects. I developed retail data pipelines and Snowflake models, and analyzed sales and inventory to identify stockout risks and support replenishment decisions.

