10101010
01010101
01010101
01010101
Architecting the Future with AI , Neural Systems & Web Technologies
Hello, I'm Tushar, an AI Engineer and Full-Stack Developer at the intersection of neural computation and web technologies. My expertise lies in training intelligent systems while building the interfaces that make them accessible.
I craft high-performance applications with ReactJS, NextJS, and TailwindCSS on the frontend, powered by NodeJS, ExpressJS, MongoDB and Django backends.
In the neural domain, I design and optimize architectures with TensorFlow and PyTorch, developing sophisticated models like LSTMs, CNNs, and custom neural networks for predictive analytics, computer vision, and natural language understanding. Beyond deep learning, I have a strong foundation in traditional machine learning, implementing algorithms like Random Forests, XGBoost, and SVMs to solve complex problems in data-driven decision-making.
Currently exploring emergent properties in large language models and optimizing for edge deployment.

Neural Skill Matrix

A comprehensive arsenal of technologies powering the next generation of AI-driven applications and intelligent systems.

Neural Activity:
Active

AI & Machine Learning

5 Technologies
Python
Python
TensorFlow
TensorFlow
PyTorch
PyTorch
Pandas
Pandas
NumPy
NumPy
Proficiency Level: Expert
[01]

Frontend Development

7 Technologies
HTML5
HTML5
CSS3
CSS3
JavaScript
JavaScript
TypeScript
TypeScript
Next.js
Next.js
React
React
Tailwind CSS
Tailwind CSS
Proficiency Level: Expert
[02]

Backend & Database

4 Technologies
Express
Express
Node.js
Node.js
MySQL
MySQL
MongoDB
MongoDB
Proficiency Level: Expert
[03]

DevOps & Tools

3 Technologies
Git
Git
Docker
Docker
Linux
Linux
Proficiency Level: Expert
[04]
System Status: Operational
Skills Matrix v2.0 | 20 Technologies Loaded

Experience_

Current

SDE Intern

LifeLabApril 2025 - Present

Working as a Software Development Engineer intern, contributing to innovative healthcare technology solutions and developing scalable applications.

Technologies

ReactNode.jsPythonAWSPostgreSQLTypeScript

Backend Developer Intern

Xoonix AIOCT 2024 - Feb 2025

Leading backend development, optimizing performance, and implementing new features.

Technologies

PythonTensorFlowDockerAWSDevOpsNest.js

Software Developer Intern

Anukash SolutionsSept 2024 - Present

Developed scalable and efficient software solutions, optimized performance, and collaborated with cross-functional teams to enhance user experience

Technologies

ReactNode.jsMongoDBGraphQL

Publications_

Published

From Classical ML to EUPN-VAE: A Unified Framework for MBTI Personality Classification

Tushar Thanvi, et al.
2024
Conference Paper

IEEE Conference

A comprehensive study presenting a unified framework that bridges classical machine learning approaches with modern EUPN-VAE (Enhanced Universal Personality Network - Variational Autoencoder) for Myers-Briggs Type Indicator (MBTI) personality classification. The research demonstrates significant improvements in classification accuracy and interpretability.

Keywords:

Machine LearningPersonality ClassificationMBTIVAEDeep Learning

Featured Projects

Exploring the intersection of artificial intelligence and modern web development through innovative solutions.

8+
Projects
5+
Technologies
Possibilities
01
LIVE

Share The Doc

A seamless document-sharing platform that allows users to upload files and instantly generate a QR code for easy access. Anyone with the QR code can scan and download the document without hassle. Built with a focus on efficiency and accessibility, this project simplifies file sharing for quick and secure distribution.

Share The Doc
Tech Stack
AIMachine LearningReactNode.js
Featured
02
LIVE

Instagram Clone

This Instagram Clone is a full-stack social media platform built with React.js, Next.js, Node.js, and Express.js, offering real-time messaging, media sharing, and dynamic content feeds. It features WebSockets (Socket.io) for instant chat, JWT authentication for security, and MongoDB with Mongoose for scalable data management. Users can upload images and videos via Cloudinary/AWS S3, receive real-time notifications, and interact with posts through likes and comments. The platform also includes stories, reels, and a dark mode for an enhanced user experience. The architecture ensures fast performance, scalability, and smooth user interactions.

Instagram Clone
Tech Stack
AIMachine LearningReactNode.js
Featured
03
LIVE

Avni PDF

A versatile tool for editing and managing PDF files efficiently, Includes Editiong, Compressing, and Converting PDF files.

Avni PDF
Tech Stack
AIMachine LearningReactNode.js
Featured
04
LIVE

Neuro Tree

NeuroTree is a AI-powered chess engine, designed to combine the strategic depth of Monte Carlo Tree Search (MCTS) with the pattern recognition strength of Convolutional Neural Networks (CNNs). By merging these two powerful techniques, NeuroTree is capable of evaluating positions intelligently, searching for the best moves efficiently, and improving its decision-making process over time.

Neuro Tree
Tech Stack
AIMachine LearningReactNode.js
Featured
05
LIVE

Sclera segmentation Using OpenCV

A computer vision-based project that segments the sclera (white part of the eye) from images using OpenCV. This technique is crucial for biometric identification, medical diagnostics, and eye-tracking applications, ensuring precise and efficient segmentation.

Sclera segmentation Using OpenCV
Tech Stack
AIMachine LearningReactNode.js
Featured
06
LIVE

Real Time Drowsiness detection System

An AI-powered system that detects signs of drowsiness in real time using computer vision and machine learning. By analyzing facial cues like eye closure and head position, it alerts users to prevent accidents, making it ideal for driver safety and fatigue monitoring.

Real Time Drowsiness detection System
Tech Stack
AIMachine LearningReactNode.js
Featured
07
LIVE

KISAN AI

n intelligent crop price prediction system powered by a Random Forest model, designed to assist farmers in making informed decisions about their agricultural produce. By analyzing historical price trends, weather conditions, and market factors, KISAN AI provides accurate price forecasts, helping farmers maximize profits and reduce financial risks.The model leverages machine learning algorithms to process large datasets, considering various influencing parameters such as crop type, location, seasonality, and market demand. This enables farmers to plan their sales strategically, ensuring they get the best possible price for their produce.Built with a user-friendly interface, KISAN AI aims to bridge the gap between technology and agriculture, empowering farmers with data-driven insights to enhance economic stability and sustainable farming practices.

KISAN AI
Tech Stack
AIMachine LearningReactNode.js
Featured
08
LIVE

Dynamic 3D Apple website Clone using Three.js and GSAP

A visually stunning replica of Apple’s website, built using Three.js and GSAP to create smooth animations and immersive 3D interactions. This project showcases advanced web animations, responsive design, and optimized performance, delivering an engaging user experience.

Dynamic 3D Apple website Clone using Three.js and GSAP
Tech Stack
AIMachine LearningReactNode.js
Featured
© 2025 Tushar Thanvi Inc. All rights reserved.