AI Researcher & Incoming MSc Student at the University of Bonn
I am an AI researcher working at the intersection of computer vision, medical image analysis, and multimodal learning. My research focuses on developing deep learning systems that can understand complex visual and linguistic information, with a particular interest in vision-language models, representation learning, and the interpretability of large multimodal models. I will be joining the University of Bonn as a Master’s student in Winter Semester 2026, where I aim to deepen my research in these areas and contribute to high-impact work in artificial intelligence.
Photo Placeholder
My research spans several interconnected areas of artificial intelligence, from visual understanding to multimodal reasoning.
Developing deep learning systems that can understand, analyze, and interpret visual information across diverse domains.
Applying AI to medical imaging problems including segmentation, classification, diagnosis, and clinical decision support.
Investigating multimodal models that bridge visual and linguistic understanding for richer AI capabilities.
Building systems capable of learning from multiple modalities such as images, text, and structured data.
Exploring language models, instruction tuning, retrieval, and the foundations of language understanding.
Developing robust and generalizable machine learning systems with strong theoretical grounding.
My research focuses on developing intelligent learning systems capable of understanding complex visual and multimodal information. I am particularly interested in applying deep learning and representation learning to challenging problems in computer vision and medical image analysis, while also exploring the capabilities, limitations, and interpretability of large multimodal models.
A selection of ongoing and completed research projects spanning medical imaging, vision-language models, and multimodal AI.
Automated assessment of cervical vertebral maturation stages using deep learning and ordinal learning approaches for clinical decision support in orthodontics.
Panjab University / PGIMER
2024 — 2025
Research investigating the failure modes, robustness, and internal mechanisms of vision-language models, including noise robustness and mechanistic interpretability.
MBZUAI
2025
Multimodal AI systems leveraging vision-language models for medical imaging and automated radiology report generation.
MBZUAI
2025
Recent and forthcoming publications in computer vision, medical image analysis, and multimodal learning.
An automated approach for assessing cervical vertebral maturation stages using deep learning and ordinal learning, enabling continuous-stage prediction from lateral cephalometric radiographs.
An investigation into the failure modes, robustness, and internal mechanisms of vision-language models, with a focus on noise robustness and mechanistic interpretability.