Rohith Vuppula — SQL Developer & Full-Stack Engineer

Who I Am

Building at the intersection of
data, systems & intelligence.

A software engineer who cares deeply about clean architecture, performant queries, and products that actually work.

SQL Developer @ Lionix LLP

Bangalore — Building enterprise-grade SQL systems and optimizing data pipelines for production applications.

April 2026 – Present

AI/ML Intern @ Nextopson

Built real-time image classification and NSFW detection models using TensorFlow.js integrated into production backend.

Jan – Jun 2025

Hackathon — Salesforce & 1M1B

Proposed AI-driven autonomous boat for river/ocean cleanup. Presented at Suit Up Global, Hyderabad.

October 2024

B.Tech Software Engineering

Balaji Institute of Technology & Science, Warangal — CGPA: 7.75

2022 – 2026

I'm a software engineer who started in full-stack web development and quickly found a passion for the data layer — specifically database design, query optimization, and making systems fast. At Lionix LLP, I write SQL that powers production applications for enterprise clients.

Outside of databases, I build ML-powered web apps, have shipped real AI models into production, and competed in sustainability-focused hackathons. I believe that well-structured data is the foundation of every great product.

13+

Technologies

9

Certifications

90%

Model Accuracy

87%

NSFW Detection

Where I've Worked

Work Experience

From AI internships to production SQL systems — here's the full picture.

● Current

L

SQL Developer

April 2026 – Present

Lionix LLP · Bangalore, India

Developing and optimizing high-performance SQL databases for enterprise web applications — including stored procedures, indexing strategies, and complex query tuning.

Ensuring data integrity, security, and compliance across multi-tenant environments through RBAC design and audit logging.

Collaborating cross-functionally with full-stack and AI/ML teams to deliver robust backend data schemas that serve both transactional and analytical workloads.

N

AI/ML Intern

Jan 2025 – Jun 2025

Nextopson

Developed and deployed a real-time image classification model using TensorFlow.js integrated into the Node.js production backend — eliminating a separate inference server entirely.

Built an NSFW content detection pipeline at 87% accuracy, directly improving content moderation and reducing manual review overhead significantly.

Applied transfer learning (MobileNetV2) with data augmentation, boosting room classification accuracy to 90% on the validation set.

What I've Built

Featured Projects

Real problems. Real solutions. Measured outcomes.

🛡️

TensorFlow.js

NSFW Shield — AI Adult Content Detector

Problem: Need for fast, private, client-side NSFW image detection without sending images to any server. Action: Built a modern web app using TensorFlow.js with MobileNet model. Added drag & drop, real-time preview, confidence bars, and scan history. Result: Fully client-side AI tool with clean light theme UI. Deployed on Vercel with excellent performance.

TensorFlow.js JavaScript HTML/CSS Express.js Vercel

GitHub Live Demo

🧠

90% Accuracy

Room Image Classification Model

Problem: Manual property photo categorization was slow and inconsistent. Action: Built MobileNetV2 transfer learning model with data augmentation using TensorFlow & Keras. Result: 90% validation accuracy — deployed live on Vercel.

PythonTensorFlow KerasMobileNetV2 Transfer Learning

Code Live

🎓

Full-Stack

Alumni Networking Platform

Problem: No structured channel for students to connect with alumni for mentorship and job leads. Action: Full-stack Django app with PostgreSQL, WebSockets for real-time messaging, and a job/event module. Result: End-to-end platform supporting live student–alumni connections with sub-100ms messaging.

DjangoPostgreSQL WebSocketBootstrap

Code Details

📊

Secure · RBAC

Internal Marks Management System

Problem: Manual grade entry was error-prone with no access control between faculty roles. Action: Node.js/MySQL system with JWT authentication, RBAC, real-time dashboards, and CSV export. Result: Eliminated manual errors, added full audit trails, and enabled live student dashboards.

Node.jsExpress.js MySQLJWTRBAC

Code Details

Technical Arsenal