[ COMPUTER SCIENCE STUDENT ]

ARUNA KUMAR

Building intelligent systems that bridge human needs and technological possibilities. Currently crafting AI solutions at the intersection of computer vision and deep learning.

View Resume
Get In Touch
0 Projects
0 Internships
0 CGPA / 10
SCROLL
01

ABOUT ME

I'm a Computer Science student at VIT Chennai with a passion for creating intelligent systems that solve real-world problems. My work spans from cutting-edge research in computer vision to building production-ready applications.

AI & MACHINE LEARNING

Deep Learning Computer Vision PyTorch TensorFlow

DEVELOPMENT

React.js Node.js Python JavaScript

RESEARCH

Camouflaged Detection Diffusion Models Attention Mechanisms Real-time Systems
Aruna Kumar
Location Chennai, India
Status Available for opportunities
02

SELECTED WORK

FEATURED
GitHub Live Demo

Agentic RAG Pipeline

An intelligent Retrieval-Augmented Generation system with agentic capabilities, featuring multi-source document processing, semantic search, and contextual response generation for enhanced AI-powered information retrieval.

Python Streamlit RAG LangChain Vector DB
Confidential

ConceptAttention Adapter

Developing an adapter module for ConceptAttention to enhance camouflaged object detection using diffusion-based cross-attention and hierarchical attention fusion.

PyTorch Computer Vision Diffusion Research
GitHub

PaveSense

Real-time pothole detection using smartphone IMU data with a hybrid CNN-LSTM model for robust mobile-based road condition monitoring.

Python CNN-LSTM Mobile ML IoT
GitHub Live

Faculty Selection System

Full-stack web application for VIT Chennai's faculty course selection with secure authentication and real-time constraint-based selection logic.

React Node.js MongoDB Full-Stack
GitHub

BumpAlert

Intelligent road bump detection system using LSTM-based sequence modeling and PPO reinforcement learning with PyBullet simulation for real-time detection.

Python LSTM Reinforcement Learning PyBullet
Confidential

Vehicle Entry System

Deep learning-based automated vehicle entry system using custom CNN-OCR for license plate recognition and RFID-based access control.

OpenCV CNN-OCR RFID IoT
03

EXPERIENCE

MAY 2025 – JUL 2025

Deep Learning Intern

IGCAR Indira Gandhi Centre for Atomic Research Kalpakkam, Tamil Nadu

Developing cutting-edge adapter modules for ConceptAttention to enhance camouflaged object detection using diffusion-based cross-attention and hierarchical fusion techniques.

  • Implemented diffusion-based cross-attention mechanisms
  • Developed hierarchical attention fusion architectures
  • Benchmarked performance against state-of-the-art models
PyTorch Computer Vision Diffusion Models Research
MAY 2024 – JUN 2024

Machine Learning Intern

IGCAR Indira Gandhi Centre for Atomic Research Kalpakkam, Tamil Nadu

Built an end-to-end deep learning-powered vehicle entry system using custom CNN-OCR pipeline for license plate recognition with RFID integration.

  • Developed custom CNN-OCR pipeline for plate recognition
  • Integrated RFID-based access control systems
  • Deployed production-ready system for campus security
TensorFlow OpenCV CNN-OCR IoT
04

GET IN TOUCH

Let's Build Something Amazing

I'm always interested in discussing new opportunities, innovative projects, and collaborations in AI, machine learning, and software development.

@
Email arundd2004@gmail.com
in
LinkedIn Connect with me
gh
GitHub Check out my code