Aya Hani

I am an AI Engineer with three years of hands-on experience building multimodal large language models (MLLMs), computer vision systems, NLP pipelines, and generative AI solutions. I focus on designing scalable ML systems and fine-tuning models to meet real-world requirements.

My projects include a multimodal driver behavior evaluation pipeline—where I fine-tuned MoViNet-A2, used YOLOv8, integrated Gemini Flash 2.5, and built scene interpretation with a visual vector database and ChromaDB—and an image-to-speech system that produces Arabic audio descriptions from images.

I continually improve models using real-time feedback and emphasize robust, data-driven solutions that deliver measurable impact. My technical toolkit spans PyTorch, TensorFlow, Hugging Face, OpenCV, Scikit-learn, and data tools such as Pandas and NumPy.

I hold a Bachelor of Engineering in Computer and Systems and have completed several certifications in generative AI, deep learning, and data science. I am motivated to join teams that push the boundaries of multimodal AI and deploy practical, user-facing intelligent systems.