LLM

RAG

Voice

Tools

Memory

Agent

n8n

API

agent = Agent(
 llm=GPT4o,
 tools=[search, code],
 memory=VectorDB
)

@workflow
def rag_pipeline():
 docs = retrieve(query)
 return llm.generate(docs)

# Voice Agent
stt → llm → tts
latency: <800ms

Hi, I'm Neloy Barman

Building Intelligent Systems That Think, Speak & Act

View My Work Get In Touch

scroll

// featured work

Projects

view all →

⭐ Featured

View Details

RAG-Based Email Query Resolution

Automatically manage your university inbox with an AI that reads student questions, matches them to your approved guidelines, and sends out friendly, consistent replies 24/7 so no one is left waiting.

AI Automation n8n OpenAI RAG LLM Self-Hosted Qdrant Docker

Source

⭐ Featured

View Details

AI-Powered Customer Feedback Automation

Automatically analyze customer feedback using AI and route it to the right channels, eliminating manual triage and enabling real-time sentiment tracking.

AI Automation n8n OpenAI LLM Multi-Modal Self-Hosted Docker

Source

Workflow

Live Demo

// latest thinking

Articles

read all →

Deep Dive into LangGraph Nodes: Inside the Core Building Block of AI Agents

Ever wondered how AI agents break down complex tasks into actionable steps? The secret lies in Nodes — the foundational building blocks that give LangGraph agents their power and flexibility.

// let's build something

Ready to work together?

Open to AI projects, agentic system design, and collaborations on cutting-edge research.

Let's Talk