PVS Prasanth I'm

About Me

I am a passionate software developer with experience in full-stack development, data analytics, and AI. I enjoy building efficient and scalable solutions, whether it's web applications, machine learning models, or automation tools. With strong problem-solving skills and a keen interest in emerging technologies, I am always looking for opportunities to learn and innovate.

Beyond coding, I have a deep passion for photography, capturing moments through my lens and exploring creativity in visual storytelling.



Experience

Data & AI Intern - Agratas, A Tata Enterprise
Jul 2025 – Oct 2025
Project - Fine-tuned a Chinese-to-English translation LLM for manufacturing, built a full-stack application with UI, database, SSO, and Docker deployment. Automated weekly EV and ESS market reports via web scraping and generated competitor intelligence reports to support leadership decisions.
Research Intern - Manipal Institute of Technology, Manipal
Jan 2025 - May 2025
Project - Built a multimodal depression prediction system by combining structured questionnaire data with Reddit-based unstructured text. Utilized sentence transformer embeddings for text and traditional ML models for structured data, integrating both using a Deep Fusion Network to enhance predictive performance and mental health assessment.
Summer of AI Intern - Swecha, Hyderabad
June 2024 - July 2024
Project - Trained, tested, and fine-tuned a Hugging Face ASR model for Telugu speech-to-text conversion. Collected diverse Telugu accents and performed data cleaning to enhance model accuracy.


Education 

Bachelor's Degree, Information Technology
Manipal Institute of Technology, 7.55 • Aug 2021 - July 2025
Intermediate, M.P.C
Aditya Junior College, 92.5% • May 2021
School, S.S.C
Aditya High School, 9.8 • May 2019


My Expertise

Application Development

Django, HTML, Cascading Style Sheets, JavaScript, RDBMS, SQL, NoSQL, MongoDB, Application programming interface

Programming & logical reasoning

Python, Java, C/C++, JavaScript, Data Structures and Algorithms, Problem Solving, Object oriented Programming

Tools & Technologies

Android Studio, Machine Learning, Artificial Intelligence, Neural Network, Semantic Web, Rapid Miner, Talend

Featured Projects

cardiopredict

CardioPredict - A Web Interface for Cardiovascular Disease

CardioPredict is a clinical decision support tool designed to assess cardiovascular risk based on anonymized medical data. It leverages a trained Random Forest model and SHAP explainability to provide interpretable predictions. The platform generates personalized PDF reports and automates result delivery via email, helping clinicians make informed decisions efficiently.

Tech Stack: Python, Scikit-learn, SHAP, ReportLab, Matplotlib, HTML, CSS, JavaScript, Django

Check it out
artgallery

A Million Drizzles - An Art Gallery Website

An Art Gallery Website that showcases and categorizes various artworks, allowing users to explore, purchase, and review paintings. Artists can upload their creations, manage their portfolio, and receive custom orders, while an admin oversees operations. This platform enhances accessibility to art and streamlines the buying process.

Tech Stack: Django, SQLite, HTML, CSS, JavaScript, SMTP

Check it out
quizzapp

Django Quiz App - An Accessible Quiz App

An Accessible Quiz App focused on inclusivity and usability, adhering to Human-Computer Interaction (HCI) principles and UI/UX design rules. It features voice commands, text-to-speech, keyboard navigation, and high-contrast visuals for enhanced accessibility. Designed with consistency, feedback mechanisms, error prevention, and user control, it ensures a seamless, intuitive, and engaging learning experience for all users.

Tech Stack: Django, JavaScript, HTML, CSS, JavaScript, SQLite, Web Speech API

Check it out
mobileprice

Mobile Price Prediction - A Mobile Price Prediction Model on Flipkart Dataset

A Mobile Price Prediction System powered by machine learning, trained on a Flipkart dataset with key smartphone specifications like RAM, ROM, processor, battery, company, and camera. Data preprocessing was performed using RapidMiner, and the model was trained using Decision Tree, Random Forest, and Naïve Bayes to ensure accurate predictions. This system helps users make data-driven decisions when purchasing smartphones.

Tech Stack: RapidMiner, Python, Scikit-learn, Decision Tree, Random Forest

Check it out
Messaging app

ConvoBuddy - A Messaging App

ConvoBuddy is an AI-powered messaging app that combines real-time chat with an intelligent chatbot for instant query resolution. Developed using Android Studio, Firebase DB, and Gemini API, it offers seamless communication, smart assistance, and real-time synchronization, making interactions more efficient and engaging.

Tech Stack: Android Studio, Java, Firebase Realtime DB, Gemini API

Check it out
yt-downloader

YT Video Downloader - YouTube Download Tool

A Python-based YouTube video downloader that enables users to download videos in various resolutions and formats using a clean, responsive web interface. Built with Streamlit and powered by yt-dlp, the tool handles errors gracefully and supports downloading playlists, audio-only formats, and more.

Tech Stack: Python, Streamlit, yt-dlp, HTML, CSS, JavaScript, Fast API

Check it out

Contact

profile_pic PVS Prasanth
pvsprasanth0209@gmail.com
Telegram
Bengaluru, India