Sonu Chaurasia

About Sonu Chaurasia

AI/ML Team Lead & Generative AI Architect with 6+ years of experience designing enterprise-scale Agentic AI and Document Intelligence platforms. I specialize in orchestrating multi-agent systems (LangGraph, MCP), building high-performance RAG pipelines, and leading cross-functional teams of 11+ engineers to deliver production-grade AI solutions for Banking & Insurance giantsCurrently, I lead the development of AgentHub, a multi-agent orchestration platform, and DocumentAI, a high-precision document intelligence engine. My focus is on combining deep technical architecture with strategic MLOps leadership to bridge the gap between cutting-edge research and scalable business value.Core Expertise: Agentic AI & Orchestration: LangGraph, LangChain, Model Context Protocol (MCP), Multi-Agent Systems. Generative AI & LLMs: RAG, Fine-tuning, Prompt Engineering, AWS Bedrock, Ollama. MLOps & Engineering: CI/CD, BentoML, Docker, Distributed Systems. Data & Search: Milvus, FAISS, SQL, Document Intelligence (OCR).Tools & Tech: Python, LangGraph, LangChain, BentoML, Milvus, AWS Bedrock, Azure, GCP, Docker, PyTorch, XGBoost, SHAP

Experience

1. Team Lead Ai Ml (Datanext) Genai, Rag & Agentic Systems

   BUSINESSNEXT

   Oct 2025 — Present · Noida, IN

   Leading AI/ML engineering initiatives within the DataNext product team, driving the design and delivery of production-grade Generative AI and Document Intelligence platforms for global banking and financial services clients.Key Responsibilities & Impact• Architected AgentHub, an enterprise-grade multi-agent orchestration platform built on LangGraph, enabling advanced agent-to-agent (A2A) communication, dynamic tool routing, and external agent integration for intelligent automation• Spearheaded DocumentAI, designing large-scale intelligent document processing systems featuring advanced OCR, template classification, AI-driven field extraction, and automated redaction for high-volume financial documents• Built production-ready RAG architectures combining sparse retrieval (BM25, SPLADE) and dense retrieval (FAISS, Milvus) to enable accurate, low-latency enterprise search across document repositories• Developed agentic AI workflows using LLMs (GPT, LLaMA), LangChain, custom tools, and Model Context Protocol (MCP) to support complex, multi-step business processes• Led cross-functional teams to deliver AI-powered capabilities for banking and insurance platforms, establishing best practices in GenAI architecture, MLOps, scalability, and secure production deployment

Education

• Himalayan University, Arunachal Pradesh

  Bachelor of Technology - BTech, Information Technology

Skills

• High Voltage
• Wiring
• Electric Power
• Electricity
• Transformer
• Motors
• Protective Relays
• Relays
• Power Systems
• Substation
• Switchgear
• Electricians
• Turbines
• Generators
• Power Plants
• Manufacturing
• Power Generation
• Power Distribution
• Electrical Engineering
• Electrical Troubleshooting