• Developed modifications to industry leading cloud CAD software name OnShape, by adding features and executing an overhaul in a fullstack development Agile SDLC
• Implemented end to end changes to TypeScript frontend (Angular, VueJS), Java server, and C++ geometry backend to introduce the new "midpointline" sketch tool, and released in live production in the latest sprint release
• Developed Unit testing and Playwright tests for my client-side code and integrated it into the CI/CD automated tests

• Collaborated with the Nutanix team to extend the NDB operator's capabilities, focusing on supporting Postgres High Availability (HA) databases using Go, Kubernetes, and Docker
• Coded adapter design OOPS pattern class to facilitate seamless bridging of client messages to backend containers in Go

• Developed Python scripts using Selenium, BeautifulSoup, Google Cloud Platform, and Pandas to collect data for 150+ Google Pixel Phone competitors and set up a MySQL database for streamlined data collection and retrieval
• Automated web crawling process and implemented error handling to reduce stoppages due to failovers by 75%

• Employed ResNet-based transfer learning to achieve an 85% accuracy rate in identifying unhealthy skin conditions
• Developed and deployed APIs on AWS EC2 instances for image segmentation models like Mask-RCNN and ResNet
• Revamped website exposing the APIs live and storing all the user-uploaded images securely in AWS S3

I developed NCSUstainability, a web application designed to predict student footfall at NC State dining halls using ARIMA time series modeling on real-world data. To achieve this, I engineered a backend using Node.js and Express.js, which facilitated RESTful APIs and spawned child processes in the Node.js environment to execute Python analysis and prediction scripts concurrently. Additionally, I integrated the ChatGPT-4 API to enhance user comprehension of predictive analytics derived from the time series models, providing users with clear and actionable insights.

I built a real-time video chat app using Node.js, Socket.io, ReactJS, and WebRTC for seamless peer-to-peer communication, audio/video streaming, and chatting through socket programming. I implemented Socket.io to ensure efficient peer connections, delivering low-latency and high-quality video calls. Additionally, I leveraged PeerJS to integrate WebRTC technology, facilitating direct and secure connections between users.

I developed a Java-based client-server chat application, similar to WhatsApp Desktop, utilizing TCP/IP protocols for communication. The backend was designed with offline user messaging capabilities, incorporating a backup functionality where messages are stored in JSON files for reliable data persistence.

I created a decentralized full-stack expense tracker using Blockchain, NextJS, and ReactJS, powered by Ethereum. I developed unit tests and staging tests for the smart contract using the Chai and Mocha frameworks, achieving 100% coverage. Additionally, I devised APIs to support user input in the form of text or speech data and utilized Postman for thorough API testing.

I designed a robust CLI database management system, creating models, relationships, and UML diagrams. Implementing the database in MySQL, I used C++ for the backend, following object-oriented programming (OOP) concepts.