Ai Agents 
Python
RAG
JavaScript
LangGraph
Open Ai
Vercel Sdk
Recent Work
Couple of my most recently built Agents
RAG Support ChatBot
🚀 Built a RAG-based AI chatbot that delivers context-aware responses using Large Language Models. Utilized Supabase (PostgreSQL + pgvector) for vector storage and semantic search, integrated OpenAI models through the Vercel AI SDK for prompt orchestration and streaming responses, and developed a responsive conversational interface. Implemented the complete RAG pipeline including document ingestion, embedding generation, vector retrieval, context injection, and LLM-powered response generation.
CLI Agent
🤖 Built an end-to-end AI agent using the OpenAI SDK with capabilities for reasoning, tool execution, and multi-step task automation. Implemented agent loop architecture, tool calling (file operations, web search, and command execution), multi-turn memory, evaluation pipelines, and observability using Laminar. Added safety mechanisms including human approval workflows, sandboxed execution, and context management through summarization and message optimization. Built with JavaScript/TypeScript, Node.js, OpenAI SDK, and Zod for reliable AI agent development
Working - MultiAgent RAG
I am currently working on a Multi Agent RAG System
Description
🤖 Multi-Agent RAG Document Intelligence System (Work in Progress) is an AI-powered document analysis platform that uses a multi-agent Retrieval-Augmented Generation (RAG) architecture to generate reliable, evidence-based reports from uploaded documents. Documents are ingested, chunked, embedded, and semantically retrieved before being processed by specialized AI agents. A Document Verification Agent validates the retrieved context, a Reflection Agent reviews and refines the reasoning, and a Report Generation Agent produces a structured final report. Built with Python, LangChain, LangGraph, and OpenAI, the system is designed with a modular architecture that can be extended with additional agents and capabilities