Vishwa

Vishwashankar T. Janakiraman

A passionate individual who always thrive to work on end to end products which develop sustainable and scalable social and technical systems to create impact.
A Master’s student in Data Science program having good experience in Machine Learning, Data Analytics and Software Development.

What I Do?

Data Science and AI:

⚡ Developing highly scalable production ready models for various deeplearning and statistical usecases.
⚡ Experience of working with Tensorflow, Keras, Computer Vision and NLP projects.
⚡ Deploying Complex quantitative modelling for time series, text and image analysis.

Full Stack Development:

⚡ Building resposive website front end using React and Angular4.
⚡ Creating application backend in Eclipse & Flask with OracleDB, MongoDB integration.


Education:

New Jersey Institute of Technology

Newark, New Jersey

Master of Science in Data Science, CGPA - 3.73/4.0
Expected Graduation: May 2021


Amrita Vishwa Vidyapeetham University

Coimbatore, India

Bachelors of Technology in Computer Science and Engineering, CGPA - 3.5/4
Graduated: April 2019



Experience:

Automatic Data Processing Inc.(ADP)

Roseland, New Jersey

Data Science Intern (Global Product & Technology)
June 2020 - August 2020

Work:

⚡ Optimized the caching strategy of a data retrieval system by using NLP techniques on unstructured network queries.
⚡ Performed extensive data analysis using Python to obtain network and user features.
⚡ Proposed and obtained additional features using Graph Clustering and Community Detection methods for inferring network and user statistics.
⚡ Followed agile methodology and improved the retrieval speed of data by considerable level.


Infosys Ltd.

Mysore, India

Software Developer Intern
December 2018 - May 2019

Work:

⚡ Developed a web application for the security department to handle complaints launched by the employees.
⚡ Incorporated OracleDB, SPRING, REST APIs in a Java backend application and Angular 4 front-end to build the web application.
⚡ Followed agile methodology and delivered the project within 80% of the proposed duration.


VoxEdu

Santa Clara (remote), California

Machine Learning Intern
May 2018 - August 2018

Work:

⚡ Developed Machine Learning models to help non-native English speakers identify pronunciation mistakes.
⚡ Incorporated data processing, feature engineering, classi cation modeling (SVM, decision trees) and neural networks (using Tensorflow) to build models in Python.
⚡ Achieved a classification accuracy of 89% and Maximum Absolute Error (MAE) of 0.45 for scoring.


National Institute of Technology

Trichy, India

Research Intern
April 2017 - April 2018

Work:

⚡ Investigated the Information Security literature to understand various security models for key sharing.
⚡ Developed security schemes using mathematical tools (prime number theory, orthogonal transformations and Eigen decomposition) for ecient secured information transfer.
⚡ Documented the ndings as Journal papers and published in International Journal of Engineering Science, Advanced Computing and Bio-Technology.

Projects

My projects makes use of vast variety of latest technology tools. My best experience is to create Data Science projects and deploy them to web applications using cloud infrastructure.

Machine Learning and Deeplearning:

Twitter bot detection using Graph Analytics

Designed a classifcation model that can predict and detect twitter bots using network analysis.
Incorporated NetworkX, Graph2Vec libraries to analyze the network graph feature embedding.

Ranking Image Captions: A Deep Learning Approach

Implemented an end-to-end Python framework system using TensorFlow and NLTK packages to rank captions by its image relevance.
Trained Convolutional Neural Network (CNN) and Word2Vec models to extract image and semantic features.
Constructed a cost function that enables Multi-modal Learning to make joint predictions.

Black-Box Adversarial Attacks on Deep Neural Networks

Implemented a black-box attack on target model and created a substitute model to learn the target model's weights.
Trained and tested the resilience of model through simulations and achieved a test accuracy of 82% on adversarial data.

Sleep Stage Prediction

Designed a model that can predict the level of sleep using EEG, EOG and EMG medical data.
Developed a LSTM time-series deep learning model to predict sleep stages and achieved an accuracy of 85%.

Smart Crop Protection System (Senior Design Project)

Designed a system that can detect intrusion of wild animals in elds to alert landlords.
Incorporated data collection, processing, and statistical models to classify animals with 92% accuracy.
Automated the image capturing process in Raspberry Pi using signals from motion sensor and developed an Android application that communicates with a server to display statistics and alerts.


Software Development:

Resource Management System

Developed an application for the carpentry workshop to manage resources and tasks eciently.
Designed and implemented a database model in MySQL and Java application to interact with user.

Cab Hiring App

Developed an Android application for cab hiring (inspired by Uber app)
Incorporated features like listing available cabs on Google maps, ride booking and payment options
Designed the MySQL database to store and access user and cab information.

Image Processing based Bottle Filling Inspection System

Developed an algorithm to detect liquid content in bottles and its level.
Adapted the algorithm to handle any liquid color and orientation.
Achieved an accuracy of 86% for detection and Mean Squared Error of 0.1 cms.

Secured Chat Application

Developed a multi-user chat application in Python using TCP sockets.
Added security feature to the communication using Cipher algorithms.

Get in touch

I am available on almost every social media. You can message me, I will reply within 24 hours. I can help you with Graph Analytics, Competetive coding, Machine Learning, Deeplearning, Aptitude, Logical reasoning and React.