Hello! I’m Townim Faisal Chowdhury, a final-year Ph.D. candidate in Computer Science at the Centre for Augmented Reasoning, Australian Institute for Machine Learning (AIML), University of Adelaide, supervised by Dr. Zhibin Liao, Dr. Johan Verjans and Dr. Vu Minh Hieu Phan.

My research focuses on explainable and trustworthy AI (XAI) for medical image analysis and multimodal foundation models. I develop methods that make AI systems transparent, reliable, and clinically meaningful that helps clinicians understand not just what a model predicts, but why. My Ph.D. thesis has led to multiple publications at top-tier venues such as CVPR, MICCAI, and ICCV.

Alongside academia, I bring 4+ years of industry experience in software and machine learning engineering. Most recently, I worked as a Research Scientist Intern at Dolby Laboratories, where I explored mechanistic interpretability in multimodal large language models (e.g., AudioLLMs) using Sparse autoencoder (SAE). Previously, I held roles in data engineering and AI product development, where I led teams, built ML pipelines, and worked closely with clients on deploying real-world solutions.

Before my Ph.D., I earned my B.Sc. in Computer Science & Engineering (Summa Cum Laude) from North South University, Dhaka, Bangladesh. There, I worked as a research assistant with Dr. Shafin Rahman on meta-learning for 3D point cloud data and with Dr. Ahsanur Rahman on graph algorithms.

My broader research interests span multimodal AI, model’s interpretability, generative AI, meta-learning, and continual learning. I enjoy bridging research and application—whether that’s developing interpretable AI for healthcare or tackling new challenges in large-scale ML systems that have real impact on human lives.

👉 You can explore my publications and LinkedIn for more details.
Currently, I am actively seeking Research Scientist, Applied Scientist opportunities, and Postdoctoral opportunities. I would love to connect and collaborate on impactful research.

For additional information, please see the CV (Last Updated on Oct 10, 2025).