Yushita Kalra

Mobile App Development (React Native)04/2025 – 05/2025

• Designed and developed a mobile art marketplace app for browsing, viewing, and purchasing curated artworks.
• Implemented OTP-based authentication using Twilio, including OTP verification and post-login navigation.
• Created a home screen featuring category filtering and a search bar for discovering artworks.
• Developed product cards displaying the artwork's name and price, linking to a detailed product page.
• Built a product details screen showing stock, description, and artwork image for each item.
• Enabled users to add items to cart, place orders, or empty the cart as needed.
• Implemented a profile screen where users can edit personal information and view their order history.
• Integrated a custom brand logo and styled the app with a personalized color palette reflecting the artistic theme.
• Planned future enhancement: Integrating a payment gateway for seamless online transactions.

Mobile App Development (React Native)01/2025 – 05/2025

• Developed a comprehensive weather application using React Native with dynamic UI components and real-time data integration.
• Implemented location-based weather tracking using Expo Location API to automatically display current conditions based on user's GPS coordinates.
• Created an intuitive temperature converter allowing users to switch between Celsius and Fahrenheit with visual feedback.
• Built a city search feature enabling users to check weather conditions in any location worldwide through the OpenWeatherMap API.
• Designed a 5-day forecast screen with detailed hourly breakdowns showing temperature, weather conditions, and humidity percentages.
• Integrated dynamic weather icons that change based on current conditions (sunny, cloudy, rainy, snowy) for enhanced visual experience.
• Engineered responsive UI components with semi-transparent elements and consistent styling for a modern, user-friendly interface.
• Implemented pull-to-refresh functionality for real-time weather data updates and state management with React Hooks.
• Utilized modular component architecture with reusable custom components for improved code maintainability and scalability.

Mobile App Development (React Native)02/2025 – 05/2025

• Developed a global news application that aggregates headlines from various countries and categories.
• Implemented a country selector allowing users to browse news from 50+ different countries, with awareness of API limitations in production environments.
• Created a drawer navigation system with categorized news sections (Sports, Business, Entertainment, Science, Technology, Health).
• Built responsive news cards displaying article title, description, author, and publication date.
• Added external linking to allow users to read complete articles in their device's browser.
• Implemented pull-to-refresh functionality and loading indicators for a smooth user experience.
• Integrated error handling for API limitations and connectivity issues with informative user messages.
• Utilized AsyncStorage to persist user's country preference between app sessions.
• Designed an engaging animated splash screen using Lottie for improved user experience.
• Future enhancement: Adding a proxy server or exploring alternative news APIs to overcome regional restrictions for international news.

Full Stack Development (MERN Stack)08/2024 – 12/2024

• Developed a task management web app with CRUD operations, priority-based task organization, and calendar-integrated deadlines.
• Implemented JWT authentication for secure login and session management.
• Designed a responsive front-end with React and built a scalable backend using Node.js, Express.js, and MongoDB.

Healthcare Management System (MERN Stack)02/2024 – 05/2024

• Created a comprehensive diabetes management platform enabling users to log vital health metrics including blood sugar levels, blood pressure, and temperature.
• Built an analytics dashboard that visualizes health data trends and generates personalized insights based on historical records.
• Implemented customized meal planning and exercise tracking features specifically designed for diabetic health management.
• Developed secure data storage for medical reports with JWT authentication to ensure patient privacy.

UCI Heart Disease Dataset01/2025 – 05/2025

• Co-authored a research paper accepted in Scientific Reports (Nature), introducing SKRB2X-Net, a hybrid deep learning model integrating XGBoost, Bi-LSTM, Kalman Filtering, and SMOTE.
• Engineered a multimodal data pipeline for structured clinical and ECG time-series data with advanced preprocessing and noise reduction techniques.
• Achieved 99.76% accuracy and 99.71% F1-score, outperforming baseline classifiers and ViT-based architectures on heart disease prediction tasks.
• Applied ensemble learning, probability averaging, and Hyperopt for robust hyperparameter tuning and improved model generalization.

Social Media Survey Dataset08/2024 – 02/2025

• Designed and conducted a survey using Google Forms, gathering responses from 530 participants across various age groups and professions. The survey comprised 22 questions exploring social media usage, mood, attention span, and sleep quality, resulting in a structured dataset for analysis.
• Applied machine learning techniques (Logistic Regression, SVM, Random Forest, XGBoost, ANN) on the dataset, utilizing PCA for dimensionality reduction. Achieved 25%-59% accuracy, uncovering key behavioral patterns linking social media usage to sleep cycle disruptions and mood swings.
• Co-authored and contributed to the research paper, which is currently submitted for review at the EBT 2025 Conference.