Srivatchan G V
Chennai, Tamil Nadu, India | +91-93425-48083 | gvsrivatchan@gmail.com | github.com/srivatchan2006 | linkedin.com/in/srivatchangv
Professional Summary

Third-year undergraduate at Shiv Nadar University, specializing in Artificial Intelligence and Data Science with a 9.42 CGPA. Passionate about building full-stack web systems and real-time intelligent applications. Eager to contribute to challenging development environments that combine automation, user-centric design, and machine learning.

Education

Shiv Nadar University, Chennai (2023 – 2027 Expected)
B.Tech in Artificial Intelligence and Data Science
CGPA: 9.42 / 10.00 (Till IV Semester)

Srimathi Sundaravalli Memorial School (2013 – 2023)
CBSE Curriculum
Class XII: 97.4%      Class X: 94.2%

Technical Skills

Languages: Python, Java, PHP, C, C++, JavaScript, SQL, HTML, CSS

Libraries & Frameworks: Django REST Framework, React, TypeScript, OpenCV, Mediapipe, Folium, LSTM

Tools & Platforms: Apache Kafka, Azure Blob Storage, MySQL, Git

Projects
Temporal Analytics for Fleet Telemetry
May 2025 – June 2025 | Apache Kafka, Azure Blob Storage, Python, Folium, LSTM

This project involved designing a robust real-time vehicle telemetry analysis system capable of ingesting, storing, and visualizing high-frequency sensor data. Apache Kafka served as the messaging backbone, continuously collecting data such as speed, fuel consumption, and engine behavior from connected vehicles.

I wrote Python-based consumer pipelines to process and clean the data in real time before storing it in Azure Blob Storage for long-term retention and scalable access. A rich dashboard was developed that visualized vehicle routes using Folium — interactive maps showed trip paths with annotations for harsh events like overspeeding or sudden braking.

A key innovation was the integration of an LSTM-based model to capture sequential driver behavior patterns. The model detected anomalies and patterns indicative of inefficient or aggressive driving styles. I also explored a reimbursement strategy for inefficient trips, where drivers could be penalized or rewarded based on metrics derived from model predictions.

Gym Membership Management System
February 2025 – March 2025 | PHP, MySQL, HTML, CSS, JavaScript

I developed a full-stack web application to handle gym operations — from member registration and login to plan management and activity tracking. The system provided role-based access where admins could onboard users, define subscription plans, and monitor attendance logs.

MySQL served as the backend storage for user credentials, plan history, and login activity. PHP handled the server-side logic for validating inputs, rendering dynamic pages, and triggering plan-related notifications. I incorporated JavaScript for client-side validation and used responsive HTML/CSS to ensure mobile compatibility.

A key feature was the implementation of check-in/check-out tracking and email-based reminders for expired plans, helping improve member engagement. The system also focused on data integrity and session management to prevent unauthorized access.

Automated Light System using Motion Detection
December 2024 – January 2025 | Python, OpenCV, HOG + SVM, Mediapipe

This project aimed to develop an intelligent lighting system that activates based on human motion and presence detection in low-light conditions. Using frame differencing and background subtraction techniques via OpenCV, the system identified significant motion in the camera’s field of view.

To ensure high accuracy, I implemented a hybrid detection model combining HOG (Histogram of Oriented Gradients) features with a Support Vector Machine (SVM) classifier to detect human figures in real time. Additionally, Mediapipe was used as a secondary validation layer to confirm human presence and reduce false positives (e.g., animals, moving curtains).

The system was configured to automatically switch lights on when motion was detected and turn them off after a configurable timeout period of inactivity. It was tested in a dark-room lab setting with impressive precision and robustness.

Hackathon Projects
AI-Powered Student Learning Web App
VIT Spectrum Hackathon – April 2025 | Django REST, React, TypeScript, Gemini AI

Collaborated in a 24-hour hackathon to build a smart educational platform that converts academic PDFs into engaging, interactive story-based content using Google’s Gemini AI. Engineered the backend with Django REST Framework to handle file uploads, AI interactions, and quiz storage. On the frontend, used React and TypeScript to create a seamless interface where students could read AI-generated narratives and take contextual quizzes. Gained hands-on experience in prompt engineering, RESTful design, and rapid full-stack prototyping under time constraints.

Positions of Responsibility

Photography Club: Core Committee Member, Shiv Nadar University (2023–Present)
Coordinated photowalks, exhibitions, and covered campus events.

Drama Club: Core Committee Member, Shiv Nadar University (2023–Present)
Participated in scriptwriting, production, and stage performance.

Extracurricular Activities

Music: Vocalist and keyboard player with stage experience.

Sports: Active football and cricket player.

Languages Known

English – Fluent  |  Tamil – Native  |  Hindi – Intermediate