Headline

Full Stack AI Engineer | Agentic AI | FastAPI · Django · Flask | LLM ·RAG ·Agents | React.js · AWS · Docker | REST APIs · MySQL · Postger sql

About

I don't just study AI - I ship it. In the past year, I've designed and deployed 4 production-grade AI systems end to end: a multi-tenant SaaS platform, an agentic software engineer orchestrator running on GCP, a clinical AI analyst with 98% accuracy, and an enterprise RAG system that cut retrieval latency by 60%. I'm a B.Tech AI/ML graduate (CGPA 8.2, JNTUK, 2025) based in Bengaluru and I build things that work in the real world, not just in Jupyter notebooks. 🔧 WHAT I BUILD ⚡ Agentic AI Systems — LangGraph, LangChain, ReAct agents, multi-agent orchestration, A2A/ACP networking ⚡ Python Backends — FastAPI, SQLAlchemy 2.0, PostgreSQL, OAuth2/JWT, async APIs at production scale ⚡ Full Stack Platforms — React.js frontends + Django ,FastAPI backends, deployed on GCP Cloud Run, Vercel, and Render ⚡ ML Pipelines — Scikit-learn, XGBoost, FAISS vector search, RAG pipelines, HuggingFace embeddings 🏆 HIGHLIGHTS - Won a 24-hour hackathon - 1st place out of 120+ teams - Built a 3-node distributed AI orchestrator on GCP Vertex AI - Achieved 98% classification accuracy on IoT botnet detection - AWS Badge Cloud Practitioner | Azure Fundamentals Certificate Udemy If you're building something ambitious with AI and need someone who can ship - let's talk. 📩 ullamnagapoojith222@gmail.com

Location

Bengaluru

AI stack

anthropic apis

claude code

codex

cursor

langchain

langgraph

Work experience

Full Stack AI Engineer at NRAI & ROBOTICS, Jun 2026 - Present · 4 mo, BTM Stage-2, Madiwala Lake, Brick building,NR Group,560076

Agentic AI Engineer - Independent Projects at Self-Employed, Jun 2025 - Jan 2026 · 8 mos, Bengaluru, Independently designed and shipped 4 production-grade Agentic AI and Full Stack systems end to end from architecture to cloud deployment. ───────────────────────────────── 🤖 Autonomous Software Engineer Orchestrator (2025) ───────────────────────────────── - Architected a 3-node multi-agent system (Orchestrator → Coder → MCP Sandbox) using LangGraph state machines and A2A/ACP networking on GCP Vertex AI + Cloud Run. - Eliminated a critical GraphRecursionError infinite loop by redesigning A2A payloads to carry last-5-node execution history resolved 100% of repeated failure cycles. Stack: LangGraph · LangChain · Gemini 2.5 Flash · FastAPI · GCP · React.js · Docker ───────────────────────────────── 🏥 Autonomous Healthcare AI Analyst (2025) ───────────────────────────────── - Built a dual-engine Supervisor agent routing between a FAISS RAG pipeline and a self-healing Pandas code interpreter cutting clinical data retrieval latency by 60%. - Achieved 98% accuracy on patient metric calculations via System Persona injection and Guardrails validation. Stack: LangGraph · FAISS · HuggingFace · FastAPI · React.js · Docker ───────────────────────────────── 🏢 Nexus AI Workspace — Multi-Tenant SaaS (2025) ───────────────────────────────── - Engineered a full stack multi-tenant SaaS platform with OAuth2/JWT auth, rate limiting (10 req/min/user), async LLM task offloading, and zero-downtime Alembic migrations. - Achieved 100% dev-to-production parity via Docker Compose — independently scaled frontend (Vercel) and backend (Render). Stack: FastAPI · React.js · PostgreSQL · SQLAlchemy 2.0 · Docker Compose · LangChain ───────────────────────────────── 🔗 Nexus LLM Gateway (2024) ───────────────────────────────── - Delivered a unified LLM gateway (Gemini + OpenAI) with automated GitHub Actions CI/CD — zero manual deployment steps on every git push. Stack: FastAPI · React.js · Gemini Pro · OpenAI SDK · Vercel · Render

Artificial Intelligence & Machine Learning Intern at Corporate Innovative Technologies Private Limited, Dec 2024 - Mar 2025 · 4 mos, Vijayawada, Andhra Pradesh, India, Completed a 4-month AI/ML internship at Corporate Innovative Technologies Pvt. Ltd., recognized under AICTE and APSCHE. 🔬 Key Contributions: - Applied supervised learning techniques including classification and regression using Python, Scikit-learn, and Pandas on real-world datasets. - Built end-to-end ML pipelines covering data collection, preprocessing, feature engineering, model training, and evaluation using accuracy, precision, recall, and F1-score. - Gained hands-on experience with NumPy and Matplotlib for data analysis and visualization of model performance metrics. 📜 Certificate ID: Ints/2025/1732774002 Stack: Python · Scikit-learn · Pandas · NumPy · Matplotlib · Machine Learning · Data Preprocessing

AI/ML Final Year Project Lead - IoT Botnet Detection System at Chalapathi Institute of Technology, JNTUK, Nov 2021 - Apr 2025 · 3 yrs 6 mos, Guntur, Andhra Pradesh, India, Led a team of 4 to design and deploy a production-grade two-fold ML system for real-time IoT network intrusion detection. 🔬 What I built: - Trained Random Forest, XGBoost, and SVM classifiers on large-scale IoT network traffic data achieving 98% classification accuracy on benign vs. botnet detection. - Owned the complete ML lifecycle: data collection, missing value handling, feature engineering on packet counts and communication patterns, hyperparameter tuning, and model evaluation using precision, recall, F1-score, and confusion matrix. - Deployed the best model via joblib into a Django MVT web app users upload a network traffic CSV and receive real-time predictions with visual benign/botnet distribution charts. 🛠️ Stack: Python · Scikit-learn · Random Forest · XGBoost · SVM · Pandas · NumPy · Matplotlib · Django · MySQL · joblib 📌 Role: Team Lead & Primary Data Scientist/ML Engineer

Education

B.Tech, Artificial intelligence and Machine learning, Chalapathi University

Public links

github: https://github.com/Unp12

nagapoojith-ai.vercel.app: https://nagapoojith-ai.vercel.app/