At Discover, I design data governance frameworks and document ingestion pipelines for compliance-driven workflows, using audit logging, Amazon S3, and AWS Lambda. I also implemented hybrid retrieval with Pinecone and OpenSearch to identify and resolve data quality inconsistencies.
At the University of North Texas, I built a Streamlit chatbot with ChromaDB to review academic content and supported faculty research with Python and SQL. Earlier, I developed forecasting and machine learning workflows at Firstsource Solutions Limited and Hinduja Global Solutions, with a focus on data quality, privacy, and model evaluation.
