Vaibhav Soni

417 Summit Ave · Arlington, TX 76013 · (512) 766-3719 · vaibhav.cs.uta@gmail.com

I'm a master's thesis student with specialisation in ML/AI and research associate in blockchain at UTA! I'm in third semester of Masters's Program in Computer Science at University of Texas.

I aspire to explore the ever exciting world of Machine Learning and BlockChain. I am currently working on BlockChain Transcation System on mobile platforms and integration of AI/ML on IoT Edge devices on clearblade platform.

Experience

Machine Learning/AI Intern

ClearBlade Inc at Austin, Texas

Increased efficiency by 68% of MQTT messages’ database insertion which is used by IOT devices to communicate. Evaluated performance of TimeScaleDB hypertables and Postgres on ClearBlade platform. Recently developed and deployed YOLO (You Look Only Once) ML model as an IoT package module on ClearBlade using JavaScript and Python. Used Agile and Scrum methods in the process of development.

May 2018 - Present

Software Developer Intern

GreedyGame Media Pvt Ltd at Bangalore, India

Categorized various Android games and their relevant ads. Manipulated UI/UX elements of the games to advertise non intrusively and increased ad views by 30%.

Oct 2016 – Mar 2017

Web Developer Intern

Agroya Soft Tech at Surat, India

Developed two web projects using WordPress, Drupal, C Panel and MySQL. Dealt with Domain registration, DNS, certificates and Apache configuration and documented the records to double the ease of operation.

July 2016 – Dec 2016

Web Developer - Team Lead

Niterider at Bangalore, India

Managed orders and deliveries for a late night food delivery startup by using HTML, CSS, WordPress, MySQL and PHP and increased the numbers of orders by launching this website.

July 2016 – Dec 2016

Education

The University of Texas at Arlington

Master of Science
Computer Science and Engineering - ML/AI and BlockChain

GPA: 3.50

May 2019

People’s Education Society-University, Bangalore, India

Bachelor of Engineering-Computer Science and Engineering

GPA: 3.53

August 2017

Skills

Programming Languages & Tools

Languages: Python, GoLang, SQL, Java, C/C++, MATLAB, Perl, Shell Scripting, R, Spark
Web Technologies: PHP, MongoDB, React, JavaScript, HTML 5.0, CSS, Node.js, AngularJS, Bootstrap, D3.js, TypeScript, CodeIgniter(MVC)
Cloud Technologies: IBM Cloud, Google Cloud, Amazon Web Services, Microsoft Azure
Tools & Concepts: Numpy, Pandas, Scikit-Learn, Keras, Git, MySQL, Tableau,RESTful APIs, Android Studio, Eclipse, Microsoft Office
Operating Systems: Mac OS X, Linux, Windows, Android

Interests

Apart from being a software developer, I enjoy most of my time being outdoors. During the warmer months here in Texas, I enjoy swimming, rock climbing, and kayaking.

When forced indoors, I follow a number of sci-fi and mystery genre movies and television shows, I am an aspiring chef, and I spend a large amount of my free time exploring the latest technology advancements in the artificial interlligence and blockchain world.

Academic Projects

  • Sentiment Analysis and Natural Language Processing for Tweets
    • Implementing a Machine Learning model to extract tokens from a ‘Tweet’ data in order to create a plot for top 25 tweets by calculating the frequency distribution using the Natural Language Toolkit in Python. Sentiment Score for each tweet would also be determined.
  • File Storage Application on Cloud Services
    • Developed Web Interface using Python-Flask on IBM Cloud, AWS, Google App Engine, Microsoft Azure for storage in containers or buckets and the retrieval of files from cloud. Visualization of data was done using D3.js and Performance was analyzed using Elastic Cache by doing random queries.
  • Detection of Brain Tumour through MRI Images
    • Used deep learning for training a model by preparing training and test sets from 1,000 MRI images. Used image processing toolbox for - Edge Mapping, Skull stripping and segmentation are applied on MRIs in preprocessing phase and then the trained model’s accuracy is computed for the targeted detection in MATLAB.
  • Analysis of Vocal Patterns to Detect Emotion
    • Used Machine Learning tools - Mel Frequency Cepstral Coefficients for feature extraction, Support Vector Machine with ‘linear’ kernel & Neural Networks for classification of 542 audio clips from Berlin database. ’Fear’ emotion was detected to assist people in stressful situations using the model developed in MATLAB.
  • Cloud Storage Server Selection based on Fault Tolerance and Depsky Architecture
    • Implemented cloud of clouds framework using Depsky Architecture where user inputs the fault tolerance values for various data hosting services and the server is selected with higher reliability. Web application was created using J2EE comprising of Java, HTML, CSS and JavaScript. Created a Git repository and published code on Github.

RESEARCH EXPERIENCE & PUBLICATIONS