• Developed evaluation metrics for hallucinations, timelines, and completeness with LLM-as-a-Judge and Chain-
of-Thought using Langchain, LangGraph, Docker, and FastAPI, increasing HopeLLM’s adoption by 45%.
• Implemented an agent-based system to automate tool calling for evaluation, collecting results, and generating
comparison analysis, saving 8+ hours of manual evaluations per deployment.
S
Schneider Electric
Software Engineer
Bengaluru, IN-KA, IN
February 2022 - present
I
Instadata
Software Development Intern
Bengaluru, IN-KA, IN
December 2021 - February 2022
C
Centre for Information Security Forensics and Cyber Resilience PES University