• Developed User Profile Settings functionality using TypeScript and React, enhancing user customization and contributing to a 100% completion of MVP features for product demo
• Delivered a responsive UI within a tight one-week sprint, supporting the startup’s client pitch and timeline goals
• Participated in daily agile stand-ups, sprint planning, and code reviews in a team of five, fostering iterative feedback and reducing UI-related bugs by 30% through peer review and continuous testing
• Implemented form validation logic and real-time visual error feedback, enhancing user experience and reducing submission errors

• Developed AI-Powered study tool using Google Gemini API and OpenCV models, winning 1st place at Demo Fest
• Coordinated the end-to-end development and integration of a real-time AI-based study assistant that reduced user distractions by 20% using gaze detection powered by Google Gemini API and OpenCV
• Designed and implemented the full-stack web application using JavaScript, HTML/CSS, and integrated front-end tracking logic with AI-based alert logic.
• Architected modular components to separate UI logic, eye-tracking models, and alert mechanisms, supporting feature expansion
• Implemented monitoring and debug logs for performance benchmarking and testing user flows under varied conditions
• Enhanced focus retention with real-time eye-tracking alerts, improving study session duration and user engagement

• Designed an AI-powered medical search tool using Llama/Ollama for fast, context-aware patient history retrieval
• Planned system architecture integrating React frontend, FastAPI backend, Epic EMR, and Elasticsearch vector search
• Evaluated open-source LLMs for secure local inference, selecting Llama for its performance, transparency, and adaptability in clinical contexts
• Defined secure data flow with patient anonymization, OAuth 2.0, and federated learning for HIPAA compliance
• Created technical diagrams, pitch materials, and Devpost content to present scalability, security, and ethical design

• Designed and built a full-stack spa booking website as part of a 4-person team, using HTML, CSS, JavaScript
• Coordinated backend system development and database integration using PHP and MySQL, implementing validation layers and SQL injection prevention to reduce potential vulnerabilities by 50%
• Implemented error handling, minimizing booking failures, and improving form submission success rate
• Collaborated in a team of four using Git for version control and feature-based branches, supporting seamless integration of frontend and backend systems

• Developed a motion-detection application using OpenCV to monitor live video feed and detect object movement, triggering automated email alerts
• Utilized multi-threading for optimized performance, enabling concurrent email notifications and image folder cleanup without interrupting video capture
• Engineered the system for low-latency notifications and minimal false positives through calibrated motion sensitivity thresholds, improving reliability for real-time security scenarios