Machine learning engineer working on document AI, interpretability, and vision-language models.

I'm Mahmoud — I build systems that read messy, real-world documents, and I care about making them reliable, trustworthy, and explainable. Right now I'm an AI engineer working on Arabic document understanding, with a research background in model interpretability. This is where I write about what I'm learning and building.

About & CV →

Latest writing