Hey, I am Niranjan. I am a sophomore at Columbia University studying Computer Science and Applied Mathematics.

I spend most of my time building backend systems, data pipelines, and projects around agents and retrieval. Recently I’ve been interested in making LLM behavior more reliable instead of purely prompt-driven.

Agent Runtime Research — Research with Professors Zhuo Zhang and Guannan Wei on a programmable agent execution environment with its own language and compiler that turns agent logic into structured workflows. The goal is to make multi-agent behavior reliable and inspectable instead of purely prompt-driven.

RAG Notes App — A full-stack note-taking app with RAG that lets you upload documents, ask questions, and get answers with citations.

SecureUs — A safety-rating web app used by 7,000+ people to check neighborhood safety.

• Agent infrastructure and reasoning systems
• Backend systems & data pipelines
• Market structure / financial systems