I build agentic AI systems that work reliably in the real world.

Aspiring AI Engineer specializing in LLM orchestration, Model Context Protocol (MCP), and production-ready document intelligence.

Selected Work

01

QUANTORA

Multi-Agent Data Analysis Platform

Query any CSV or Excel dataset in plain English — no SQL required. Orchestrates Planning and SQL Agents over MCP to generate schema-aware DuckDB queries.

Python · MCP · DuckDB · Prompt EngineeringView Project →
02

System Diagnostics Agent

Secure AI Troubleshooting with HITL

An AI agent that diagnoses Windows system issues from natural-language reports. Enforces human approval before any repair, producing full audit logs.

Python · LLM Agents · HITL DesignView Project →
03

Doc2Ops

Document Intelligence Pipeline

Converts customer Sales Order PDFs into structured data. Combines deterministic feasibility logic with LLM-driven planning — eliminating hallucinated outputs.

Python · Docling · LLM OrchestrationView Project →

About

I'm an AI engineering student at Government College of Technology, Coimbatore, graduating in 2027.

My work centers on the gap between "it works in a demo" and "it works when it matters" — whether that's enforcing human-in-the-loop approvals before repairs, eliminating hallucinated assumptions in manufacturing outputs, or building modular agents that are genuinely extensible.

Certified in Anthropic MCP and Oracle Cloud Infrastructure, I'm looking for AI engineering internships to contribute to production LLM systems.

AI / LLM Engineering

Multi-Agent Systems, MCP, Agentic Workflows, Prompt Engineering, Document Intelligence

Languages & Databases

Python, Java, SQL, DuckDB, MongoDB, MySQL