Hi, I'm Ayush Patel!

I'm a Computer Science student at Georgia Tech, focusing on AI and Information Internetworks. Over the years, I've had the opportunity to work on various exciting projects that have shaped my understanding of technology.

During my internship at Meta, I worked on improving caching systems for their Content Distribution Network, where I gained invaluable insights into large-scale infrastructure. Additionally, through the Startup Exchange Fellowship, I contributed to an AR application using Apple Vision Pro, aimed at providing real-time captions for the hard of hearing.

My past experiences also include data engineering at State Farm and astrophysics research at MIT Haystack Observatory, where I studied molecular clouds. These diverse roles have shown me how technology can solve real-world problems across different domains.

Currently, I'm a Teaching Assistant for Computer Systems & Networks at Georgia Tech, which allows me to support fellow students while deepening my own understanding. I'm always eager to learn and contribute to projects that push the boundaries of technology.

If you're passionate about innovation or curious about tech, I'd love to connect and explore potential collaborations or simply engage in insightful discussions.

Professional Experience

Meta

Software Engineering Intern (C++)

May 2024 – Aug 2024

  • Enhancing caching mechanisms for Facebook's Content Delivery Network to improve hit rate
  • Developing real-time hotness tracking system to monitor and manage content popularity dynamically

State Farm

Software Engineering Intern (Apache Spark/Scala)

May 2023 – Aug 2023

  • Utilized CI/CD to engineer distributed data pipeline to process over 100,000 daily transaction events, employing AWS Glue to manage Apache Spark nodes
  • Implemented advanced optimization techniques like predicate pushdowns, shuffle tuning, and broadcasting, resulting in a 23% reduction in data processing costs and a 14% boost in job performance

MIT Haystack Observatory

Research Intern

Jun 2021 – Aug 2022

  • Devised & documented efficient method for computing kinematic distances; improved estimate to within 0.5 kpc
  • Presented documentation at Northeast Star & Planet Formation conference

Recent Projects

Green Loops (React/Node.js)

Developed a platform that allows students to sell used books easily and receive instant payment. The system streamlines the buying process by managing inventory, pick-up schedules, and payments.

Sports Arbitrage Finder Bot (Python)

Created a bot that identifies arbitrage opportunities across 13 sports betting sites, generating 2-4% in returns. Optimized performance with multithreading and deployed it using Docker on Amazon EC2 for continuous operation.

EmpowAR - Assisted Hearing (Swift/Python)

EmpowAR is an innovative augmented reality application designed to assist individuals with hearing impairments. Utilizing Apple Vision Pro technology, it provides real-time captions for spoken words in the user's environment, enhancing communication and accessibility. The project combines Swift for the AR interface and Python for advanced speech recognition algorithms, demonstrating a powerful intersection of cutting-edge technology and social impact.

Astronomical Data Analysis Dashboard (Python)

Displays velocity dispersion, integrated intensity, and other moment maps for a specific molecular cloud.