# Sujeet Gund
Agentic AI Engineer · M.Tech AI @ VIT Bhopal
Building autonomous agentic systems, production RAG pipelines, and LLM-powered infrastructure.

## About
Engineering intelligent systems from first principles to production.
Final-year Integrated M.Tech in AI at VIT Bhopal, currently building production AI systems at Divam Technologies.
I specialize in agentic architectures — LangGraph multi-agent workflows with HITL checkpoints, hybrid RAG with pgvector, and FastAPI backends on Google Cloud Run. The kind of systems that handle real workloads, not just demos.
Outside work, I ship ambitious side projects: an email processing SaaS, an anonymous geo-social platform, and tooling that pushes LLM orchestration further than most tutorials go.

## Contact & Links
- Location: Barshi, Maharashtra, IN
- Email: sujeetgund@gmail.com
- LinkedIn: https://linkedin.com/in/sujeetgund
- GitHub: https://github.com/sujeetgund
- Website: https://sujeetgund.in
- Resume: https://sujeetgund.in/resume

## Experience
- GenAI Developer Intern at Divam Technologies (Apr 2026 - Present): Sole architect of Publie.ai, an AI-powered LinkedIn automation platform — built the FastAPI backend, LinkedIn OAuth + publishing API integration, BullMQ scheduled posting, and deployed the full stack on AWS. Also delivered a production LangGraph chat agent with pgvector RAG for an enterprise client.

## Projects
- GroundedAI — Self-Correcting Multi-Source Agentic RAG: A self-evaluating agentic RAG platform powered by LangGraph, Reciprocal Rank Fusion (RRF) pgvector + tsvector hybrid search, and local zero-cost DeBERTa v3 NLI faithfulness scoring to eliminate hallucinations (94.2% grounding precision, <45ms CPU NLI latency). (LangGraph, FastAPI, pgvector, Python, Next.js 16, Cross-Encoder DeBERTa, Redis, Tavily API) - https://github.com/sujeetgund/grounded-ai - https://sujeetgund.in/projects/grounded-ai
- rzp Merchant — AI-Native E-Commerce & Agentic Commerce (MCP/ACP): An autonomous e-commerce platform implementing Model Context Protocol (MCP) and Agentic Commerce Protocol (ACP), allowing external AI agents (Claude, Cursor, ChatGPT) to discover, basket-optimize, and execute transactions via Razorpay under 3-tier security mandates (100% spending mandate enforcement, 3.8x co-purchase conversion uplift, <120ms MCP tool latency). (Next.js 16, LangGraph.js, Gemini 2.5 Flash, Razorpay SDK, MCP Protocol, Drizzle ORM, pgvector, Better Auth) - https://github.com/sujeetgund/rzp-merchant - https://sujeetgund.in/projects/rzp-merchant
- MailMind — Agentic Email Processing & HITL Orchestration Engine: An event-driven email processing agent built with LangGraph and Resend Inbound APIs. Features intent classification (96.5% extraction accuracy), FAISS RAG auto-drafting, and PostgreSQL-persisted Human-in-the-Loop (HITL) approval checkpoints yielding an 88% operational time reduction. (LangGraph, FastAPI, Resend Webhooks, ChatGroq, FAISS, PostgreSQL, Next.js 16, Docker) - https://github.com/sujeetgund/mailmind - https://sujeetgund.in/projects/mailmind

## Skills
- Languages & Databases: Python, TypeScript, PostgreSQL, MongoDB
- AI & ML: LangGraph, LangChain, RAG Systems, pgvector, PyTorch, NLP, Agentic AI
- Backend & Infrastructure: FastAPI, Next.js, Docker, BullMQ, AWS, GCP, GitHub Actions

## Education
- VIT Bhopal University: Integrated M.Tech in Artificial Intelligence (Sep 2023 - Mar 2028), CGPA: 9.31

## Certifications
- Complete Data Science, Machine Learning, DL, NLP Bootcamp (Udemy, Dec 2025)
- Applied Machine Learning in Python (University of Michigan, Dec 2024)
- Supervised Machine Learning: Regression and Classification (DeepLearning.AI, Jul 2023)
- AWS Cloud Essentials (Amazon Web Services (AWS), Jun 2023)