Extended a RAG application with client-in-context workflows across 5 specialized sub-agents, improving response relevance for nearly 200 financial advisors. Also led migration from ChromaDB to Elasticsearch and from a Python RAG stack to an agentic framework built on Google ADK.
Architected a Spring Boot and AWS Lambda-based offset calculator that processed around 600K records and reduced runtime from nearly 2 hours to about 8 minutes. Enhanced RAG experiences for financial advisors, modernized ETL pipelines, and cut infrastructure costs by migrating workloads from EKS to ECS.
Optimized an internal query-language-to-SQL parser using Spring Boot, lowering response latency by about 25 percent. Added Databricks SQL support and built nested retrieval with Apache Calcite, improving query reliability and retrieval speed.
If you think you might have an opportunity for me, please feel free to reach out to me on the contact information provided below !