Data Analyst with a strong quantitative background and an academic trajectory toward Artificial Intelligence and Machine Learning. Experienced in data analysis, statistical methods, Python-based data processing, and data-driven problem solving. Interested in machine learning, graph/network analysis, and computational methods for complex datasets. Seeking a research internship at KAUST to develop practical research experience in graph machine learning and network analysis.
Used a large driving behavior dataset (speed, acceleration, braking, time-of-day, distance driven) with several million rows. Cleaned and transformed the data, handling missing values, outliers, and inconsistent records. Engineered features such as speeding ratio, harsh braking per 100 km, night driving percentage, and average trip duration. Trained logistic regression and gradient boosting models to predict the probability of an accident in the next 30 days. Evaluated models using ROC-AUC, precision-recall, and calibration plots. Built a Power BI dashboard to visualize risk scores by driver, vehicle, region, and behavior patterns, including filters to simulate improvements in driving style. Documented the full workflow in Jupyter notebooks and a clear project README.
Built a custom web app integrating the Claude API to summarize lecture notes, generate interactive practice quizzes, and answer student queries in real time. * Authored clean, modular code with robust API rate-limiting handling and published setup documentation on GitHub.
Organized and facilitated hands-on coding workshops, partnered with faculty to promote events, and led a team to deploy a project management portal.
Spearheaded a hackathon, coordinated participant teams, and served as a technical mentor.