Alif Ba! is an Arabic character recognition system that utilizes a convolutional neural network and a gradient-based optimization machine learning algorithm. Its purpose is to identify individual Arabic alphabet characters while disregarding other character forms like initial, medial, or final. Although Alif Ba! is not flawless, it continuously improves over time.

The underlying implementation of Alif Ba! relies on the Adam optimization algorithm, which serves as an alternative to the conventional stochastic gradient descent method. This algorithm adapts to estimate lower-order moments, making it computationally efficient, requiring minimal memory, invariant to diagonal rescaling of gradients, and well-suited for handling large-scale problems involving substantial amounts of data and parameters. The optimization algorithm updates the network weights iteratively using training data and eventually finds the optimal weight after multiple iterations. The neural network governs the iterations, with forward propagation updating the weights and back propagation optimizing those weights.

MIVVI Shop is an ecommerce platform for selling beauty products. I designed and programmed the entire web app with help from third party open-source libraries. Involved in backend, front-end, database administration, and deployment to Heroku.

Centipe Text Editor is a fully-functional text editor designed as a free alternative to the proprietary Notepad software available on Windows platforms. It has a wide range of features, including line numbering, text formatting, and search and replace. It also supports a variety of file formats, including plain text, HTML, CSS, JavaScript, and Python.