At Hero FinCorp, I build core services for the ELM platform, including the Banking Analysis microservice that orchestrates AA journeys, consent collection, and BRE decisioning for pre-approved loan limit enhancements.
I own roughly 80% of the pre-qualified loan journey end-to-end across backend and frontend. I also proposed an API orchestration sequence that reduced third-party spend per customer by about 57% and implemented circuit breakers for stronger downstream resilience.
I've built conversational AI prototypes using LiveKit, TTS/STT, and LLM orchestration, alongside AI products including an NL-to-SQL BI platform and a Hybrid RAG document intelligence platform. My work includes self-healing SQL agents, semantic caching, and Redis session memory that reduced repeated-query latency by 98%.
Previously at Mapsense, I built location-query and spatial-analysis capabilities including heatmaps, clustering, and buffer analysis. I work across Java, Spring Boot, React, TypeScript, AWS, databases, and AI agent frameworks.
