koushal95 / websiteArchive

Personal Website

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About Me

I'm Prudhvi Ogirala, a computer science graduate with focus in Data Science and Machine learning. My interests lie in Data Science, Story-telling with visualization, and Web development. Feel free to connect with me on Linkedin.

Education

  • Masters in Computer Science with focus in Data science.

    Coursework: Machine Learning, Artificial Intelligence, Deep Learning, Probabilistic Models, Intelligent Data Analysis, Cloud Computing.

  • Bachelors in Computer Science and Engineering.

Work Experience

  • Data Analyst Intern (University Outreach Program) at JP Morgan Chase and Co. (Jan 2017 - June 2017)

  • Web Developer Intern at Ixtlesoft (March 2016 - July 2016)

Projects

  • Amazon's EC2 Spot Instance Price Prediction

    Amazon web services provides different pricing models pay-per-use, fixed, and auction-based (spot price). It is seen that the spot price is a minimum of 5 times cheaper than the other pricing models but there is no guarantee that you will be given the instance. It depends on the price you bid. Thus, analysis of historical data for spot price in order to efficiently (minimal cost) schedule the jobs is important.

    Designed and built a model to predict EC2 instance spot price, which enables to schedule EMR jobs efficiently, to reduce the cost incurred to bare minimum. Used Amazon’s price history data and developed the model in Python with results pertaining to error rate less than 1%.

  • Rest Weather API

    Built a REST API in Python using Flask to serve the weather data of Cincinnati and deployed it on cloud. Built a model to forecast the weather data based on the historical data, accounting to trends and seasonality of the time series data. The results of the model are made available in the API. Used google charts to display the forecast plot for the next seven days.

  • Optical Character Recognition

    Performed classification using Support Vector Machine, Logistic Regression, Decision Tree classifier, and Naïve Bayes classifier to recognize handwritten English alphabet. The work can be extended to diverse applications like assistive technology to blind, automatic processing of invoices, etc.

  • Hearing aid for Deaf using Holograms

    Developed a mixed reality application using C#, Unity and Windows SDK, to assist the deaf in communication with motivation to provide seamless communication. The application eliminates the cost of a language translator. Collaborator: @ranjit95

  • Internal Grade Automation

    Designed and developed a web application, using Javascript, MySQL and PHP, for college to automate the management of sessional examinations, aimed at saving time and money. Collaborator: @ranjit95

Skills

  • Data Science

    Machine Learning, Scikit Learn, TensorFlow, Prophet, Pandas, Numpy, SQL, Python, Matlab, Weka.

  • Data Visualization

    Tableau, D3.js, Matplotlib, Seaborn, Google Charts, High Charts.

  • Web Development

    Python, Flask, Angular JS, HTML5, Javascript, Jquery, Bootstrap, PHP.

  • Other

    Data Analysis, Data Mining, AWS, Git, Docker, MS Excel, Unity3D, Java.

Contact

Feel free to contact me at koushal.ogirala@gmail.com and drop in a connection request on Linkedin.

About

Personal Website


Languages

Language:CSS 59.1%Language:HTML 40.9%