I am a Ph.D. student in Computer Science at the University of Illinois Urbana-Champaign, advised by Dr. Ismini Lourentzou, with an expected graduation in May 2026. My research focuses on trustworthy and multi-modal machine learning, with interests in vision-language models, grounded response generation, and conversational AI. I specialize in building robust AI models that can handle missing information, mitigate adversarial attacks, and defend against harmful or malicious content.

My experience includes internships at Google DeepMind on the Gemini App team and three summers at IBM Research (Almaden Lab). My work at Google DeepMind resulted in one paper submitted for internal review and one invention disclosure. Previously, my work at IBM Research on human-in-the-loop learning and PII detection resulted in a conference paper, a granted patent, and two additional invention disclosures.

I am the co-lead of the winning team for the 2025 Amazon Nova AI Challenge, where we developed models that are robust against malicious prompts and reliably generate code secure against Common Weakness Enumerations (CWEs). Additionally, I was part of a finalist team in the Amazon Alexa Prize TaskBot Challenge 2, contributing to task-oriented conversational AI, dynamic caching, and harmful content filtering.

Previously, I earned my M.S. in Computer Science from Virginia Tech and my B.Sc. in Computer Science & Engineering from the University of Dhaka. In addition, I have experience working as a Machine Learning Engineer at TigerIT Bangladesh Ltd. and a Full Stack Developer at Enosis Solutions. My expertise spans large language models, vision-language models, conversational AI, and robust ML, with strong programming skills in Python, C, and full-stack development.

News

• August, 2025: Our paper on adversarial robustness of code language models got accepted in EMNLP ‘25.
• July, 2025: We won the Amazon Nova AI Challenge.
• May, 2025: Started my internship at Google DeepMind.
• February, 2025: Our paper on part-focused semantic co-segmentation with vision-language models got accepted in CVPR ‘25.
• May, 2024: Started my internship at IBM Research.
• October, 2023: My paper on multi-modal representation learning for image-to-text and text-to-image retrieval got accepted at the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV).
• August, 2023: Our team advanced to the finals of the Amazon Alexa Prize Taskbot Challenge 2.
• August, 2023: My paper on human-in-the-loop model selection for set expansion got accepted at the 32nd ACM International Conference on Information and Knowledge Management (CIKM).