About

From communication systems to machine learning.

Telecommunications taught me to look at the whole system: the signal, the network, and what can fail between them.

I studied communication and electrical engineering at Beirut Arab University, then saw those ideas in practice during a telecommunications internship at OGERO. Working with access networks, fiber, and mobile systems gave me a practical view of how engineering decisions are shaped by reliability and physical constraints.

My UAV GPS spoofing project brought me into machine learning through a familiar problem: deciding whether a signal can be trusted. A football forecasting project taught me a different lesson about time-aware data, leakage, and the uncertainty that remains even after careful modeling.

At the American University of Beirut, I moved deeper into how models learn. My master’s thesis studied adaptive optimization under differential privacy and how training decisions affect underrepresented groups.

What comes next

The kinds of problems I want to keep working on.

Problems that interest me

I’m drawn to systems that must make useful decisions from noisy or sensitive data, especially when accuracy alone hides security risks, privacy costs, or uneven outcomes across groups.

Where I’m headed

I’m interested in applied machine learning and ML engineering roles where I can work across modeling, experimentation, and real-world system constraints. I want to build models that work under those constraints, evaluate them beyond a single headline metric, and understand where and why they fail.