My research focuses on developing advanced vision-language models and agentic AI systems with applications in climate science, healthcare, and infrastructure. I explore multimodal learning approaches to create scalable AI solutions with measurable real-world impact.
Developing multimodal architectures that bridge visual and linguistic understanding for complex reasoning tasks.
Designing autonomous agents capable of goal-directed behavior and multi-agent collaboration.
Exploring joint representations across vision, language, and other modalities for robust understanding.
This survey systematically reviews the emerging paradigm of agentic AI, analyzing its potential to transform computing environments through autonomous decision-making and multi-agent coordination.
Presents a novel deep learning framework for glacial lake outburst flood (GLOF) risk assessment using multimodal satellite imagery and climate data.
Survey paper "Governed By Agents: A Survey On The Role Of Agentic AI In Future Computing Environments" published as preprint on arXiv.
Paper accepted at FIT 2024 on using deep learning for GLOF risk reduction in Pakistan's northern mountain ranges.
Joined Cohere for AI (C4AI) community as an active member focusing on computer vision and multi-agent systems.