Hello, I'm

Ranjith Kumar

A Software Devloper & AI Enthusiast.
I learn, develop, and optimize AI models for mobile use.

About Me

My journey in the digital frontier, driven by curiosity and a passion for building the future.

I am a passionate and driven developer currently in my final year of B.Tech in Computer Science at Kumaraguru College of Technology, with a deep focus on leveraging Artificial Intelligence and Machine Learning to solve real-world problems. I specialize in building intelligent systems that blend strong technical foundations with thoughtful, user-centered design.

My experience spans frameworks like PyTorch and tools such as Transformers and LangChain, enabling me to work confidently at the intersection of research and application. I have a particular knack for understanding the fundamentals behind complex models, which helps me not only implement AI systems but also dissect and optimize them for better performance.

Recently, I was honored with the Best Process Excellence Award at my college for a project that showcased innovation, impact, and technical depth — an achievement that reflects my commitment to building effective and efficient solutions. I also have experience deploying models and developing full-stack ML applications, giving me end-to-end visibility of the development pipeline.

While I enjoy development, my core interest lies in AI research — exploring model behavior, challenging existing limits, and uncovering novel use cases. In the next few years, I aspire to grow as an AI researcher, product engineer, or startup founder, creating solutions that are not only scalable and reliable, but also socially meaningful and accessible across domains like healthcare, education, and sustainability.

My Core Toolkit

Python

TensorFlow

PyTorch

Scikit-learn

Docker

Git

My Journey & Education

Milestones and roles that have shaped my expertise and perspective.

B.E in Computer Science

Kumaraguru College of Technology

2021 - 2025

Specialized in Artificial Intelligence and Machine Learning, completed a final year project on an Intelligent Healthcare Assistant, and received the Best Process Excellence Award.

AI Intern Trainee

Infosys Springboard

OCT 2024 - DEC 2024

Designed and implemented a system for detecting fraudulent signature verifications using:

Edge Detection
Contour Analysis
Pixel-Based Matching

My Creations

Here's a glimpse into some of the exciting projects I've worked on, showcasing my skills and passion for innovation. Each project represents a unique challenge and a step forward in my journey.

Explore All Projects

Intelligent Healthcare Assistant

A comprehensive AI solution designed to assist healthcare professionals and patients:

Chatbot Module: Interactive Q&A based on patient prescriptions
Computer Vision Module: Extracts and interprets prescription details
Voice-Enabled Module: Disease prediction assistance

MLPythonHTMLCustom LLM

GitHub Documentation

Remote Sensing using ML

Conducted an in-depth study using satellite-based remote sensing data. Achieved third place in an inter-university competition for geospatial data processing.

CNNDataFlowPython

.

OULAD Student Performance

Analyzed student performance patterns using the Open University Learning Analytics Dataset. Implemented machine learning models to predict outcomes.

Data Eng.ANNScikit-learn

.

My Publications

Authored works and research contributions that I've had the privilege to share with the community.

Medical Assistance System Using Large Language Model and Medical Data Management

Ranjith Kumar P, Dinesh Arun M, Soundararajan N, Nivetha R, Kaviya Priya S, Nivethika R.

Published In: 2025 3rd International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA) Location: Coimbatore, India Date: 04-05 April 2025

This project aims to create a multi-module system to deal with issues related to healthcare personal record keeping and communication. There are three modules working synchronously. The process of extracting meaningful information from the prescriptions and storing them in a database is done in Module 1. In Module 2, a fine-tuned Large Language Model is implemented with information retrieval capabilities that enables users to ask medical-related queries and retrieve their medical history using natural language. With the help of the voice interaction technology included in Module 3, users can interact verbally with the system by giving voice instructions, and the system will respond accordingly.

View on IEEE Xplore Cite DOI: 10.1109/ICAECA63854.2025.11012384

Insights & Reflections

Sharing knowledge, experiences, and thoughts on the evolving tech landscape.

Reinforcement Learning

Mar 7, 2025

Been a busy few weeks, but I’m back! Lately, I’ve been deep into Reinforcement Learning—mainly Q-learning—and exploring attention mechanisms in LLMs. Also building an RL-powered Tic-Tac-Toe game, nearly done, just ironing out training issues. Found a great video on attention too—more on that soon!

Empowering Patients with Technology

Feb 24, 2025

Healthcare can be overwhelming, so I built a smart interface—both as a webpage and Android app—to help patients manage documents and get personalized assistance. In this post, I break down the tech, workflow, challenges, and what’s next, repeating key parts to make sure it all clicks.

How I learnt Deep Learning and LLMs

Feb 23, 2025

Sharing my journey from deep learning basics to exploring LLMs and diving into reinforcement learning. It’s been late nights, rented GPUs, and a serious YouTube deep dive. If you're curious about attention, LLMs, or why RL has me hooked—let’s break it all down, step by step.

View All Posts

Have an idea? A challenge? Or just want to talk tech?
Let's make something amazing together.

Drop Me a Line