Sujeet Gund

Agentic AI Engineer · M.Tech AI @ VIT Bhopal

Building autonomous agentic systems, production RAG pipelines, and LLM-powered infrastructure.

Barshi, Maharashtra, IN
Resume
Sujeet Gund

About

Engineering intelligent systems from first principles to production

15+Advanced AI Projects
9.33University CGPA
5+RAG Pipelines Built

Final-year Integrated M.Tech in AI at VIT Bhopal, with hands-on experience 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 deployed on AWS and GCP. I focus on building systems that handle real workloads, not just demos.

Outside work, I ship ambitious side projects: an agentic email processing SaaS, an AI-native commerce engine, and open-source tools that push LLM orchestration beyond tutorial limits.

Work Experience

GenAI Developer Intern

Divam TechnologiesDivam Technologies•Remote
Apr 2026 - Sep 2026

Led end-to-end GenAI development across real-time voice infrastructure, multi-agent systems, and enterprise RAG pipelines deployed on AWS.

  • Built a hybrid RAG system combining vector search with file-based lookup to improve factual precision on structured queries and reduce retrieval latency by 10x (50-100 ms to <5 ms).
  • Engineered an async user context pipeline that processes resumes, PDFs, and portfolio links using rolling summarization to compress ~15k raw tokens into structured 2-3k token context windows.
  • Developed a WhatsApp voice agent using Deepgram, OpenAI, Cartesia, and LiveKit WebRTC with sentence-level speech streaming and Redis caching, cutting time-to-first-audio from ~4s to <700ms and lookup silence to <1s.
  • Configured production AWS infrastructure with GitHub Actions CI/CD across isolated dev/prod environments, integrating LinkedIn OAuth, Razorpay payments, and automated WhatsApp notifications.

Featured Projects

01

GroundedAI

Self-Correcting Multi-Source Agentic RAG

Eliminates LLM hallucinations across complex multi-source documents by combining hybrid RRF vector search with local NLI faithfulness verification (94.2% grounding precision).
LangGraphFastAPIpgvectorNLI DeBERTa v3Next.jsRedis
Read MoreSee Code
02

rzp Merchant

Autonomous Agentic Commerce Engine (MCP/ACP)

Implements MCP enabling external AI agents to autonomously discover products, optimize baskets, and execute Razorpay transactions under multi-tier security mandates.
Next.jsLangGraphGemini 2.5Razorpay SDKMCPpgvector
Read MoreSee Code
03

MailMind

Agentic Email Processing & HITL Engine

Automates inbox operations with intent classification (96.5% accuracy), RAG draft generation, and PostgreSQL-persisted Human-in-the-Loop approval checkpoints (88% operational time reduction).
LangGraphFastAPIResendGroqFAISSPostgreSQL
Read MoreSee Code

Skills

Languages & Databases

PythonTypeScriptPostgreSQLMongoDB

AI/ML

LangGraphLangChainRAG SystemspgvectorPyTorchNLPAgentic AI

Backend & Infrastructure

FastAPINext.jsDockerBullMQAWSGCPGitHub Actions

Certifications

Building RAG Agents with LLMs

Issued by NVIDIA

View Credential

Complete Data Science, Machine Learning, DL, NLP Bootcamp

Issued by Udemy

View Credential

Applied Machine Learning in Python

Issued by University of Michigan

View Credential

Supervised Machine Learning: Regression and Classification

Issued by DeepLearning.AI

View Credential

AWS Cloud Essentials

Issued by Amazon Web Services

View Credential

Education

Integrated M.Tech in Artificial Intelligence

VIT Bhopal University

Sep 2023 - Mar 2028
CGPA: 9.33