I am a wolfpack student pursuing my Masters in Computer Science at NC State with track in data science. My recent experiences include working with aws automated deployment with CICD pipeline and working on django+python framework. My past experience also include working on data science projects in the fields of sentiment analysis and computer vision.
-
Experience
Recently I have been working on a web-based platform with USDA as a part-time software developer where my roles include the development of web applications showcasing plant and human disease spread across through various factors through visualizations and deploying them through containerization on aws and implementing CICD pipeline. I also improved Scalability of the application using Elastic Load Balancer and AWS EC2 Auto Scaling.
-
Projects
Software Developer | USDA – North Carolina State University Jul 2020 – Present
• Developed web application showcasing the data related to various diseases in plants and animals using R-Shiny and Python-Django. • Containerized application and designed continuous delivery/deployment using AWS CodePipeline.
• Improved Scalability of the application using Elastic Load Balancer and AWS EC2 Auto Scaling.
Research Assistant | Laplante Labs – North Carolina State University Feb 2020 – Present • Studied and developed automated computer vision-based solutions using python for cell structure and cell division study.
• Devised Auto Clustering of proteins over frames of microscopic images using Density based clustering (DBScan) with 96% accuracy.
Data Science Intern | E2excel Dec 2018 – May 2019 • Created social media analytics module which included Twitter and Facebook sentiment analysis for the Information Intelligence
Platform using LSTM, Python and Django. The sentiment analysis achieved an overall accuracy of 81.22% on test data. • Devised algorithm for question tree generation using clustering and POS tagging to find parent-child relationship.
• Developed a module leveraging data analysis to interpret the survey response data and show demographics.
Python Developer | First Align Solutions Jun 2018 – Dec 2018 • Processed stock market data and converted it to various technical indicators which increased readability of the patterns in them.
• Researched and constructed models for technical indicators to be utilized for predicting the future stock values via Neural Network. • Parallelized the operation of generating technical indicators on real-time data.
• Developed rule-based model for automated trading using interactive brokers and python.
Computer Vision, Deep Learning
• Modeled a novel approach of predicting sheep's weight from its image through segmentation and neural network-based regression model to minimize the human involvement in measuring weight of the sheep in python.
• Presented at “International Conference on Artificial Intelligence and Data Engineering 2019” to be published in Springer series • Published in “Advances in Artificial Intelligence and Data Engineering”. [ https://doi.org/10.1007/978-981-15-3514-7_4 ]
Python, Apache Spark, GraphFrames
• Formulated a project to find the important members of terrorist organization to target, using Spark GraphFrames and the algorithm utilized graph properties like degree, centrality and articulation point.
Natural Language Processing, Machine Learning
• Used BERT architecture to classify emails from Enron email Dataset provided by CMU containing 5 lakh emails of 150 users.
• Facilitated grouping of similar emails and most common discussion topics
• Accepted at “Future Technologies Conference 2020”, to be published in Springer series “Advances in Intelligent Systems and
Computing”.