Projects
As a passionate programmer, I have worked on several exciting projects that have honed my skills in software development, machine learning, and data analysis.
​
Here are some of the projects that I have worked on
Apps (iOS + Andriod)
Tracker and Geolocator App - Android Studio + Google Maps SDK
I developed a tracker and geolocator app using the Google Maps SDK and Android Studio. The app allows users to track and locate specific points of interest or individuals in real-time.
By leveraging the power of the Google Maps API, the app provides accurate and reliable location data, enabling users to easily navigate and find their desired destinations.
The project involved integrating the Google Maps SDK into the Android Studio development environment, implementing functionalities such as displaying maps, tracking user location, and providing directions.
Decentralized Finance App
I developed a decentralized finance (DeFi) web application, which aimed to provide users with a secure and transparent financial ecosystem.
The application utilized blockchain technology to eliminate intermediaries and facilitate peer-to-peer transactions. I designed and implemented smart contracts using Solidity, ensuring the integrity and efficiency of financial operations. The web application offered features such as decentralized lending, borrowing, and trading, allowing users to interact with various digital assets. I also integrated decentralized identity solutions to enhance security and privacy.
​
Social Networking App
During my project, I developed a social networking app using Android Studio. The app aimed to provide users with a platform to connect and interact with each other.
I leveraged the capabilities of Android Studio to design and implement an intuitive user interface, allowing users to create profiles, post updates, share media, and engage in conversations with other users. The app incorporated features such as real-time messaging, notifications, and user authentication to ensure a seamless and secure user experience.
Throughout the development process, I focused on creating a visually appealing and user-friendly interface, optimizing performance, and ensuring the app's compatibility across different Android devices and versions.
Machine Learning iOS App for Image Classification using Create ML and Core ML (Apple Machine Learning Frameworks)
Leveraging the power of Apple's advanced machine learning frameworks, Create ML and Core ML, I developed an innovative solution for image classification on iOS devices.
Powered by Core ML, the app performs all image classifications directly on the user's device. This on-device processing ensures privacy, responsiveness, and functionality even without an internet connection. Users can easily classify images captured in real-time or from their gallery with a simple tap, thanks to the intuitive user interface.
Full Stack Web Development
Task Management Application
The Task Management Application is ideal for individuals, teams, or businesses seeking a centralized solution to manage tasks efficiently. Its customizable features, ease of use, and real-time collaboration capabilities make it a valuable tool for enhancing productivity and achieving organizational goals. The app can be deployed locally and can also be connected with the AWS or Azure Suite.
​
Frontend: React.js, Redux (for state management), Bootstrap or Material-UI (for UI components)
Backend: Node.js, Express.js, RESTful APIs
Database: MongoDB (or any other preferred database)
Authentication: JSON Web Tokens (JWT)
Payment Gateway: PayPal
Cloud Services: Amazon Web Services (AWS)
Real-time Communication: Socket.io or Firebase Realtime Database (for messaging system)
Version Control: Git, GitLab
Deployment: Docker containers for scalability, Continuous Integration/Continuous Deployment (CI/CD) tools for automated deployment
Tensorflow based Machine Learning Projects
Recommender Systems with Matrix Factorization
This project focuses on building a personalized recommendation system using matrix factorization techniques. By analyzing user-item interaction data, the model learns the latent features of users and items. Matrix factorization methods help in capturing underlying patterns in the data, enabling the system to make accurate predictions about user preferences. This type of recommender system is widely used in online platforms to suggest movies, products, or content tailored to individual users' tastes and preferences.
Anomaly Detection with Autoencoders
Anomaly detection is crucial for identifying rare and abnormal patterns within datasets. Autoencoders, a type of neural network, are employed in this project to learn the inherent structure of the input data. During training, the model learns to encode normal data patterns accurately. When exposed to anomalous data, the reconstruction error is significantly higher, allowing the system to flag these instances as anomalies. This technique finds applications in fraud detection, network security, and quality control, where identifying unusual patterns is essential.
Object Detection with Tensorflow Object Detection API
Object detection is a fundamental computer vision task involving locating and classifying objects within images or video frames. This project employs TensorFlow's Object Detection API, a powerful tool that facilitates the creation of robust object detection models. By leveraging pre-trained models and fine-tuning specific datasets, the system can accurately identify multiple objects and their respective classes in real time. This technology is pivotal in applications like autonomous vehicles, surveillance, and augmented reality, enhancing object recognition capabilities.
Image Classification with CNNs
Image classification using CNNs is a foundational deep-learning task that assigns labels to images based on their content. In this project, a CNN model is trained on large datasets containing labeled images (CIFAR-10, ImageNet) to recognize intricate patterns and features within the images. The convolutional layers enable the network to capture spatial hierarchies, making it highly effective for image recognition tasks. This technology is widely used in applications such as medical imaging, facial recognition, and content moderation, enabling automated and accurate image categorization.
Data Science Project with PowerQuery and Excel
Sales Analysis and Forecasting Dashboard
The Sales Analysis and Forecasting Dashboard, created using Excel's PowerQuery and advanced Data Science techniques, provides real-time insights into sales trends and future predictions. By cleaning and integrating raw sales data, the aim was to build an interactive interface where stakeholders can explore historical patterns, identify customer preferences, and forecast demand. This tool not only optimizes inventory management but also guides strategic decisions, enabling businesses to stay ahead of market dynamics.