Backend & Applied AI/ML Engineer | Computer Science Student | Innovation Pioneer
Driven by curiosity and code, I specialize in building fault-tolerant systems, AI-powered applications, and cutting-edge research solutions.
I'm Vatsal Gupta, a Computer Science and Engineering student with a strong inclination toward software development, AI, and problem-solving. I enjoy exploring real-world challenges through technology and continuously work on sharpening my technical and analytical skills.
Over time, I've worked on various projects and gained experience through internships that helped me build a practical understanding of how technology can be applied to create impact. I'm adaptable, self-driven, and focused on evolving with the fast-changing tech landscape.
With every challenge I take on, I aim to grow as a professional who's not just technically capable but also reliable, curious, and committed to continuous improvement.
Current Mission: Pursuing B.Tech in Computer Science and Engineering at Amity University (CGPA: 8.66) while building innovative solutions in AI/ML, backend systems, and research-grade applications.
Specializations: Backend development for fintech, privacy-first offline AI systems, research-grade ML projects, and fault-tolerant data processing systems.
Worked with the Technology team on "Java Payment Gateway: Settlement & Reporting" project. Gained hands-on experience in backend development for financial systems, contributing to critical modules related to data integration, reporting, and secure transaction workflows.
Actively contributed to community outreach and development initiatives focused on education, healthcare, and social welfare under the Amity Internship Program.
A sophisticated Python framework that orchestrates specialized AI agents for comprehensive text analysis. Features include summarization, sentiment analysis, named entity recognition, and 7 other NLP tasks. Built with thread-safe communication, adaptive trust scoring, and consensus-based result aggregation. Zero external dependencies, pure Python implementation.
Built an offline Retrieval-Augmented Generation system using Meta's LLaMA 3.2 and FAISS for fast vector-based document retrieval. Enables natural-language questions with contextually grounded answers from uploaded documents (PDFs, DOCX, TXT). Fully offline system ensuring data privacy and user control, ideal for academic, legal, and enterprise use cases.
Research-oriented framework to benchmark deep learning optimizers across various architectures and datasets. Analyzes performance, efficiency, stability, and generalization with robustness testing under label noise. Automated logging with Weights & Biases and generates LaTeX-ready tables and comparative plots for academic publishing.
Machine Learning-based IDS using UNSW-NB15 dataset with RandomForest classifier, hyperparameter tuning, and model explainability through LIME. Achieved improved detection metrics with visual insights via confusion matrix, permutation importance, and local feature explanations.
Real-time computer vision project that recreates the famous invisibility cloak effect using color detection and background replacement. Optimized with HSV color masking and morphological transformations for real-time webcam feed processing.
Built a machine learning model to classify text into positive, negative, and neutral sentiments using Python and NLTK. Applied advanced text preprocessing techniques and feature engineering with TF-IDF and Bag-of-Words models.
Fully responsive Amazon-inspired e-commerce frontend showcasing core web development skills. Recreated key pages including homepage, product listings, search functionality, navigation header, and footer with modern design principles.
Created a series of interactive games and graphical interfaces using Python's Turtle module. Built engaging mini-games including obstacle dodgers, click-based puzzles, and animations demonstrating creativity and game logic implementation.
Pursuing B.Tech in Computer Science and Engineering with strong technical foundations in programming, software development, and machine learning. College has been instrumental in shaping my interest in AI, full-stack development, and problem-solving through hands-on learning.
Completed secondary and senior secondary education with focus on Physics, Chemistry, Mathematics, and Computer Science. Maintained academic excellence while actively participating in sports, especially badminton and cricket.
Issued: July 18, 2025 | Country: India | Class: 24-01
This registered design pertains to an AI-Enabled Skin Disease Identification and Diagnosis Robot, developed as a collaborative innovation in the field of medical technology. The system integrates artificial intelligence, computer vision, and robotic automation to analyze skin conditions and assist in early-stage diagnosis.
The design was filed under Class 24-01 with the Indian Patent Office and focuses on the physical and functional attributes of a robotic unit capable of interpreting dermatological symptoms through image processing and machine learning models.
Innovation Impact: This work represents a convergence of healthcare and engineering, contributing to the development of accessible, AI-driven diagnostic tools in dermatology. The project involved a multidisciplinary team combining expertise in software, hardware, and clinical insight.
Contributed as a Hospitality Volunteer during CONFLUENCE-2025, the 15th International Conference on Cloud Computing, Data Science & Engineering. Supported guest coordination, session logistics, and on-ground hospitality management for speakers, panelists, and delegates from 20+ countries.
Successfully participated in HaBuild's Largest Online Yoga Camp on International Yoga Day 2025. Joined 7.52 lakh people from 169+ countries in a massive global wellness event focused on daily yoga practice, breathing techniques, and personal well-being.