About Me
Development Stats
Background
Education
MS Computer Science • University of Colorado Boulder
BTech IT • DJ Sanghvi College • 8.52 CGPA
Research
8 Research Papers • International Journals
Patent Applied • ML Scheduling Algorithms
Leadership
Vice Chairperson • DJ Init.AI Club
Research Intern • USC & IIT Patna
Building the future with AI & Innovation
Work Experience
January 2025
January 2025
AI/ML Research Intern
University of Southern California (USC)
January 2025 - April 2025
Los Angeles, CA
Contributed to ML models for detecting emotional disturbances and cognitive impairments by analyzing smartphone usage patterns, GPS, and environmental data (air quality, humidity, allergens)
Built and optimized data pipelines, reducing processing time by 60%, and performed advanced data analytics
Visualized insights on a web-based dashboard to support real-time mental health monitoring and behavioral analysis
December 2024
December 2024
Web Development Freelancer
Freelancing
December 2024 - January 2025
Remote
Designed and developed a dynamic property listing website with advanced search capabilities and filters
Implemented user-friendly features, including property-related news and an intuitive interface
Optimized website functionality to deliver seamless navigation with comprehensive property details
September 2024
September 2024
Research Intern
Indian Institute of Technology (IIT) Patna
September 2024 - December 2024
Patna, India
Conducted comparative analysis of Vision-Language Models (VLMs) and Large Language Models (LLMs)
Assessed model effectiveness in interpreting region-specific sports data from India and France
Employed rule-based, image-based, and history-based methodologies to evaluate model performance
June 2024
June 2024
AI/ML Intern
Datamatics
June 2024 - August 2024
Mumbai, India
Built Agentic Process Automation (APA) systems using Langchain, Autogen, CrewAI, LaVague, Azure and OpenAI
Developed an Adaptive Capital Allocation System with LLMs and RAG for real-time investment insights
Created OCR-based Automated Document Processing tool with advanced preprocessing for low-quality images
Improved text extraction and decision-making accuracy by 70% using RAG and LLM-driven analysis
December 2023
December 2023
Web Development Intern
Centre for Development of Advanced Computing (CDAC)
December 2023 - May 2024
Mumbai, India
Developed an interactive virtual dice simulator using ReactJS, Material UI, JavaScript, Plotly, and CSS
Enabled real-time probability exploration through simulations and live graph updates
Integrated dynamic Plotly charts to visualize theoretical vs. experimental outcomes
Enhanced student engagement and conceptual understanding through intuitive UI and smooth animations
Skills & Expertise
Full-Stack Development
Building scalable applications from frontend to backend with cutting-edge technologies.
AI/ML Engineering
Developing intelligent systems using deep learning, NLP, and computer vision techniques.
DevOps & Cloud
Architecting robust cloud solutions with automated deployment pipelines.
Research & Publications
Contributing to cutting-edge research in AI, ML, and computer science domains.
Data Science
Extracting insights from complex datasets using advanced analytics and visualization.
Performance Optimization
Building high-performance systems that scale efficiently under heavy loads.
Research & Publications
Adaptive Web Accessing Tool for Visually Impaired People with Explainable AI
International journal Educational Administration: Theory and Practice
Built a multilingual voice assistant for the visually impaired with personalized content delivery using ML and NLP.
Detection of Fake Online Products Using Unsupervised GAN with Grad-CAM Visualization
Presented at BIDA 2025; to be published in Springer’s “Smart Innovation, Systems and Technologies” series.
Developed a GAN-based framework with Grad-CAM to detect fake online products, achieving high accuracy and explainability.
Comprehensive NLP System for Research Paper Discovery and Similarity Analysis
International Conference on Communication, Computing, and Data Security (ICCCDS 2025)
Built an NLP-based system to compare research papers, extract key entities, predict citation impact with LSTM, and rank relevance by abstract similarity.
Enhanced Nighttime Crime Surveillance Using Federated Learning: A Technical Evaluation of Aggregation Methods
4th International Conference on Advances in Data-driven Computing and Intelligent Systems (ADCIS 2025)
Federated learning for crime detection with 90-95% accuracy across aggregation methods.
Resource-Aware Scheduling Algorithms For Machine Learning
Patent Application
Novel scheduling algorithms for ML workloads in heterogeneous environments.
AdversaMark: Robust Digital Watermarking through GANs
2nd International Conference on Sustainable Computing and Intelligent Systems (SCIS 2025)
Generative adversarial networks for robust digital watermarking solutions.
Optimized ECG-Based Detection of Hypertensive Heart Disease Using Federated Machine Learning
5th IEEE ASIANCON 2025
Federated machine learning approach for ECG-based hypertensive heart disease detection.
Enhancing Hypertensive Heart Disease Detection Through ECG Signal Optimization
7th International Conference on Communication and Intelligent Systems (ICCIS 2025)
Developed an image-to-signal ECG pipeline using ML/DL for automated hypertensive heart disease detection from paper-based records.
Featured Projects
Get In Touch
Let's Connect
I'm always interested in discussing new opportunities, innovative projects, and collaborations in AI/ML, software development, full-stack development, and research.