Amruth Nare

Innovator, Engineer, Student, and Investor


My name is Amruth Nare and welcome to my personal portfolio website. I am a highly motivated Blockchain and Machine Learning enthusiast looking to bridge my knowledge of Finance, Business, and Computer Science together.

I am a junior at the University of Maryland: College Park and am currently completing a double major in Computer Science and Finance. At the University, I am a member of the Business, Society, and Economy scholars program, a specialized and accelerated living program designed to emphasize entrepreneurial skills and leadership through seminars, colloquiums, and business projects.

From a young age I have been captivated by software and how it can be used to better the lives of everyday people as well as allow multinational corporations to function. Along with programming, I am also passsionate about investing, finance, and management. In the future I would like to pursue a career in a venture capital or Fintech firm that specializes in blockchain and machine learning technology, a perfect combination of my interests!

Other than programming and investing, some of my interests include running, weight lifting, learning about economics, and playing video games.


Software Engineer Intern

  • Interned at GEICO on the Mobile Cloud Integration Tier team
  • Streamlined calls to the Mobile Cloud API by migrating endpoints to serverless functions hosted on Azure using C#
  • Reduced latency in calls to Redis cache by 300% and implemented secure callback functionality in C Sharp
  • Designed and Implemented a test driven automated diagnostic tool with edge cases for production server performance by analyzing relational data using Logistic and Linear Regression which increased anomaly detection by 15%
  • Utilized Splunk, Python, and SciKitLearn to query, preprocess, analyze, and test API calls on non-production servers


Software Engineer Intern

  • Interned at the Capital One Center for Machine Learning with the Document Intelligence and Vision Engineering team (DIVE)
  • Built and trained Machine Learning models using Paddle to perform Object Detection on ID documents
  • Created a Docker image for the DIVE team to utilize when training Paddle machine learning models and published to Artifactory
  • Adapted the models to create an extensive end-to-end pipeline that utilizes PP-YOLOv2 and PP-YOLOE object detection algorithms for any general object type
  • Utilized Python, Paddle, Docker, AWS S3, and Artifactory


Research Intern

  • Interned under Dr. Zhang over the summer at the University of Maryland
  • Identified a lack in efficiency in computing power when performing machine learning
  • Created a Machine Learning architecture that leveraged data parallelism technology to perform efficient distributed ML
  • Wrote an extensive research paper outlining the architecture
  • Achieved a 67% increase in efficiency when training various ML models across 4 nodes
  • Built with Horovod, TensorFlow, AWS DLAMI EC2, and OpenMPI


Medius Developer

  • Contracted out by Open Water Accelerator to a financial research startup
  • As the sole back end developer, created a REST API for the Medius web application with NodeJS, ExpressJS, ReactJS, OAuth2.0
  • Managed the user and post database by implementing MongoDB Atlas clusters
  • Deployed the MVP with AWS Route 53, EC2 instances, VPC and GoDaddy


Project Manager: Automated Form Filler

  • Developed an automated form filling electron application to aid teachers in filling out redundant field trip forms. Managed the entire workflow with an agile scrum based development process
  • Programmed and built the front end logic and GUI for the App to process PDF & DOCX forms and output queries for any necessary information
  • The application removes the repetitiveness in field trip form filling and is dynamic to year to year changes in MCPS forms
  • Returned the completed field trip forms in a Zip containing PDFs
  • Built with ElectronJS, Express/NodeJS & Bootstrap


SourceAmerica Design Challenge: ChangeCounter

  • Worked with Furnace Hills Coffee, a cafe that employs people with developmental to develop an application to help people freely operate the cash register
  • Created a mobile application that worked fluently with the Square Point Of Sale system to help employees return change from cash transactions
  • The App displays the amount of change on the screen with counters and implements machine learning to verify the employee is handing back the proper amount of change
  • Allowed employees to operate the cash register independently and accurately
  • Built the iOS app with Swift, Apple's CoreML, YoloV3 model, and the Square-POS API