Mehrab Tanjim joined Adobe Research as a Research Scientist in April 2023. His research focuses on query understanding and disambiguation, synthetic data generation for robust model training and evaluation, automated optimization of LLM-based agents for reasoning, planning, and decision-making tasks, as well as trust, bias, and content moderation in generative AI systems. His work spans a variety of applications, including question answering, image generation, image-to-image translation, and text-to-image synthesis.

Mehrab earned his Ph.D. in Computer Science from the University of California, San Diego (UCSD), where he was advised by Prof. Garrison W. Cottrell. His doctoral research explored methods to enhance the fairness and robustness of generative models across diverse modalities and tasks. During his Ph.D., he also interned at Adobe Research multiple times (Summer 2020, 2021, and 2022) where he focused on debiasing image generative models. Prior to doctoral studies, he received his bachelor’s degree in Computer Science from the Bangladesh University of Engineering and Technology (BUET), where he worked on scalable machine learning algorithms in distributed environments.

More information about Mehrab’s research and publications can be found on his website. For academic collaboration and internship opportunities, please reach out at tanjim [at] adobe [dot] com.