youssef safwat

I am a Computer Science student focused on AI with hands-on experience building ML pipelines, annotation workflows, and SAP ABAP solutions. I bring a collaborative, adaptable approach and a track record of delivering reliable models and production-facing tools.

I have engineered end-to-end ML projects including CNN-based image classification achieving 87% test accuracy and a Transformer-based Arabic-English translator using Hugging Face models for robust bilingual performance. I design preprocessing, training, and validation pipelines to ensure model generalization and semantic fidelity.

In applied settings I developed an AI-driven league scheduler using Particle Swarm Optimization that met all constraints, and I created an SAP HR control panel using ABAP, ALV reporting, and SAP screens to streamline employee workflows. I prioritize maintainability and clear documentation when delivering features.

Through internships and QA roles I drove annotation quality to 95% accuracy and reduced labeling errors by 20%, supporting more reliable model training and faster iteration. I seek roles where I can contribute technical rigor, mentorable growth, and cross-functional collaboration.