I'm
Abdullah Ibrahim
AI/ML Engineer
base in Somewhere.

The Attendance System with Face Recognition automates attendance tracking by identifying individuals through facial features captured by a camera. It uses computer vision and deep learning (e.g., CNNs with OpenCV or FaceNet) to detect and recognize faces in real time. Once recognized, the system logs the person's attendance in a database, ensuring accuracy, reducing manual errors, and preventing proxy attendance. Project Link.

This Streamlit app performs customer segmentation using the K‑Means clustering algorithm on user-uploaded datasets (or a default one). Users select two numeric features (e.g., age, income, spending score), determine the optimal number of clusters via the Elbow method, and visualize distinct customer groups in a dynamic scatter plot with autosummaries and profile tables. It enables businesses to uncover meaningful segments—like high-income big spenders vs. cautious budget shoppers—and tailor marketing strategies accordingly. Project Link.

Project Link.

Project Link.

This Streamlit app uses an XGBoost classifier to predict a person's obesity level based on inputs like age, gender, height, weight, and lifestyle habits. Trained on a labeled dataset, the model achieves high accuracy in classifying obesity categories (e.g., underweight, normal, overweight, obese). The interactive interface allows users to input their data and receive real-time predictions, along with visual explanations of the model’s behavior. Project Link.

This Streamlit app predicts the risk of heart failure using machine learning models trained on clinical data such as age, blood pressure, cholesterol, and ejection fraction. Users input their medical details, and the model instantly provides a risk assessment—helping identify individuals who may require further medical attention. The tool aims to support early diagnosis and preventative care by leveraging data-driven predictions. Project Link.

This AI Translator uses TensorFlow and LSTM-based sequence-to-sequence models to translate text from one language to another. The encoder LSTM processes the input sentence and compresses it into a context vector, which the decoder LSTM then uses to generate the translated sentence word by word. The model learns linguistic patterns and grammar rules from bilingual datasets, enabling accurate and fluent translations. Project Link.

This AI Text Generator uses Long Short-Term Memory (LSTM) networks to learn patterns in text data and generate coherent, human-like sequences of text. By training on large datasets such as books, articles, or song lyrics, the model captures the context and structure of language over time. Once trained, it can generate new sentences or paragraphs by predicting the next word or character based on previous inpu Project Link.