About Me

I'm a meticulous data scientist who's deeply passionate about problem-solving. I'm always fueled by curiosity and on the hunt for new discoveries. There's something about diving into data, unraveling patterns, and pushing the limits of what we know that just gets me going. And that drive doesn't stop with me—I thrive in team settings, leveraging my leadership skills and constantly expanding my network. With every challenge I tackle, I'm growing and evolving in this ever-changing world of data science. Keep Scrolling !

Skills

Experience

Education

  • Data Science
    Analyzing complex data to derive insights and make informed decisions.
  • Data Engineering
    Developing multi-tier data engineering solutions to solve evolving data problems
  • Data Analysis
    Create user-friendly data visualization reports and dashboards
  • Computer and Machine Vision
    Build Deep learning models to solve image-processing based problems
  • Data Science Intern
    DevelUP, Bangalore, India  

    July 2023 - Aug 2023

  • ML Intern
    Suvidha Foundation  

    Mar 2023 - Apr 2023

  • ML Intern
    Veg Route 

    Feb 2023 - Mar 2023

  • AI/ML Intern
    Atsuya Technologies  

    Jan 2022 - Apr 2022

  • M.S Data Science
    Northeastern University

    Expected May 2025

  • B.E Electronics and Communication Engineering
    Anna University

    July 2019- May 2023

Certificates

Microsoft Certified

Azure Data Engineer

Through this certification, I acquired expertise in implementing efficient partition strategies, executing comprehensive data exploration, crafting queries utilizing SQL serverless and Spark cluster, and ensuring data security while monitoring storage and processing. I thoroughly enjoyed creating numerous pipelines utilizing Azure Data Factory, Synapse Analytics, Azure SQL Server, and Databricks.

Verify Credentials

Research Work

Advanced Semiconductor Classifiers Using Machine Learning Techniques

This study proposes an innovative method utilizing logistic regression and random forest classifiers for advanced semiconductor classification, addressing the growing complexity and importance of semiconductor testing in the electronics industry.

XGBoost based Prediction and Evaluation model for Enchanting Subscribers in Industrial sector

In this paper, we aimed at leveraging XGBoost to enhance customer engagement and address the dynamic nature of subscriber behavior, crucial for optimizing revenue generation in service management

Academic Projects

Image Processing based Animal Intrusion Detection system in Agricultural Field using Deep Learning

Image processing-based animal incursion detection system in agricultural fields using Raspberry Pi and YOLOv7, emitting repellent sounds and sending SMS alerts upon intrusion to prevent crop damage and human-wildlife conflict efficiently.

Azure - Lakehouse Architecture Project

This project utilizes Azure Self Hosted Integration Runtime to extract and stage data from local machines into a data lake, transforming it through Databricks for PowerBI visualizations.

A Gesture-based Tool for Sterile Browsing of Radiology Images

A vision-based hand gesture recognition system enabling sterile manipulation of radiology images in real-time, enhancing surgical workflow and interaction within electronic medical record databases.

ICU Management System

This system efficiently gathers, records, and manages comprehensive patient data, facilitating real-time monitoring and informed medical decision-making.

REAL ESTATE MARKET CLUSTER ANALYSIS

Segmenting housing markets for tailored investments and marketing strategies, driving corporate success through targeted growth opportunities.

MNIST Digit Recognition with Supervised Learning

This project explores various supervised machine learning algorithms for classifying handwritten digits on the MNIST dataset. We compare and analyze the performance of different models, including Bayes rule classifiers, autoencoders, neural networks, and more.

TWEET ANALYSIS

Utilized scikit-learn's CountVectorizer and LogisticRegression, along with NLTK's PorterStemmer, to achieve 94% accuracy in categorizing tweets, showcasing expertise in text preprocessing and machine learning.

BOSTON HOUSE PREDICTION

Employed various machine-learning models to predict housing prices in Boston, including Linear Regression, Decision Tree Regression, and Random Forest Regression, to provide actionable insights for homeowners and real estate professionals

Contact Me

  gnanasekar.o@northeastern.edu content_copy

  ovi13ya@gmail.com content_copy

View/Download Resume
 
Success Your message has been submitted