Hi There,
I'm Varun Mehta
I am into
About Me
I am a Full-Stack developer based in Los Angeles, USA. I am currently pursuing a Master's in Computer Science at USC, with a focus on AI-driven applications. I have experience building web applications, AI tools, and real-time systems. I love exploring new technologies and building scalable systems that make a real-world impact.
email : varunjay@usc.edu
place : Los Angeles, USA
Education is not the learning of facts, but the training of the mind to think.
University of Southern California (USC)
CGPA: 3.71/4.0
Coursework: Design and Analysis of Algorithms, Database Management Systems, Machine Learning, Web Technologies, Natural Language Processing, Autonomous Cyberphysical System, Deep Learning.
Dwarkadas J. Sanghvi College of Engineering
CGPA: 9.38/10.0
Coursework: Artificial Intelligence, Data Warehouse and Mining, Advanced Algorithms, Distributed Computing, Big Data Analytics, Operating Systems, Computer Networks, Information Security.
Developed a stock market application using the MERN stack for the web version and a native iOS app with SwiftUI. Features real-time stock data visualization powered by Finnhub API and Highcharts.
Built a robust image classifier for waste management using ResNet50v2 and CNN architectures. The model classifies waste into six categories, aiding efficient recycling.
Implemented a data extraction and summarization tool for medical reports using AWS Textract and GPT-2. Enhanced accuracy in identifying structured data, aiding healthcare insights.
Developed an end-to-end pipeline to extract and summarize ESG reports using OCR and advanced NLP models. Achieved high summarization accuracy for streamlined ESG data analysis.
Developed an adaptive cruise control system integrating deep learning models with simulation environments to enhance autonomous driving capabilities.
Developed a cloud-native Java application with an automated CI/CD pipeline using AWS CodePipeline, Docker, and ECS. Streamlined build, test, and deployment processes.
June 2024 - October 2024 | Atlanta, GA
September 2023 - Present | Los Angeles, CA
May 2024 - August 2024 | Remote
November 2021 - April 2022 | Mumbai, IN
This paper presents a novel approach to generating concise summaries from video transcripts using Transfer Learning techniques. The model was trained on a dataset of 1000 diverse video samples, achieving a Route-1 score of 0.43, and is particularly effective for applications in content summarization.
This paper discusses the development of a voice recognition personal assistant for local businesses to improve operational efficiency and customer service.
This paper presents a novel enhancement to the RSA cryptosystem, improving both security and efficiency in cryptographic applications.
