Kunal Mahatha

Hi! I’m Kunal Mahatha, a PhD student at École de technologie supérieure (ÉTS) and the International Laboratory on Learning Systems (ILLS), working under Prof. Christian Desrosiers and Prof. José Dolz. My research focuses on adapting vision models for downstream tasks—particularly semantic and open-vocabulary segmentation—spanning training-free transformer-based methods and efficient state-space architectures (SSMs).

I bring a mix of industry and research experience across AI, computer vision, and machine learning. In industry, I have worked at PharVision Capital, contributing to AI Agents, RAG pipelines, and LLM-based systems, and at Sony India Software Centre, where I gained experience in large-scale software and applied ML workflows. On the research side, I have been a Mitacs Globalink Research Intern at ÉTS Montréal, pursued a research internship at the Computer Vision Center (CVC) in Spain, and worked as an undergraduate researcher, where my interest in vision and learning systems first began.

In my free time, I enjoy getting outside during the summers or cooking something good at home. I’m also a great fond of watching movies, especially when I need inspiration or a break from research. Alongside this, I have a growing passion for content creation as a way to give back to the community by sharing knowledge and experiences.

I’m always open to collaborations, discussions, and new challenges. Feel free to explore my work and reach out—would love to connect!