Researchers studying large-scale artificial intelligence, microbial biomanufacturing and causal inference methods are among the six Cornell assistant professors who recently received National Science Foundation Faculty Early Career Development Awards.

Each will receive a minimum of $400,000 over a five-year period from the program, which supports early-career faculty “who have the potential to serve as academic role models in research and education and to lead advances in the mission of their department or organization,” according to the NSF. Each funded project must include an educational component.

Recent recipients from Cornell:

• Mohamed Abdelfattah, electrical and computer engineering (Cornell Tech and Cornell Engineering), will use his award to further his research on improving the efficiency of large language models (LLMs), which feature a memory footprint that can extend to terabytes of data and requires immense computational capabilities. This project will first focus on automated partitioning and mapping algorithms, as these form the foundations by which LLMs can be deployed and optimized on both existing and new distributed computing platforms. He will then investigate the codesign of new hardware systems and algorithms to boost the performance of large-scale deep-learning models.
• Saleh Kalantari, human centered design (College of Human Ecology), will use NSF funding to improve human-environment interaction in public spaces, aiming to ensure people reach their destinations on time and with minimal stress. The economic and human toll of navigational difficulties is often underestimated, and researchers have found tremendous economic losses resulting from poor architectural design, measured in millions of missed appointments and flights, and other issues. The goal of this project is to use data from actual human navigational experiences – including that of older adults, an often-overlooked group – to help improve computational design-evaluation tools.
• Sijin Li, Smith School of Chemical and Biomolecular Engineering (Cornell Engineering), will use her award to further her work on engineering microorganisms via synthetic biology methods. Microbial biomanufacturing is a powerful alternative approach to producing bioactive plant natural products. Engineered microorganisms can produce bioactive plant natural products in a short period of time by fermentation in closed vessels, thus providing an efficient approach to strengthen the supply chain. This work will also develop educational and research activities for women students, including summer programs, interdisciplinary education, research training and mentorship.
• Karan Mehta, electrical and computer engineering (Cornell Engineering), will further his work on quantum systems based on trapped atomic ions, particularly chip-integrated hardware platforms that enable scalable and stable delivery of light-to-ion arrays. The award supports his group’s investigations of ion-light interaction in tailored spatial field profiles, in which fine spatial variations of the field profile can potentially enhance basic operations relevant for quantum computation and atomic clocks. The work aims to explore new atom-light interactions enabled in these configurations, potentially opening a new frontier for quantum control in scalable platforms.
• Francesca Parise, electrical and computer engineering (Cornell Engineering), will focus on the analysis and control of multi-agent systems involving heterogenous interactions among strategic agents – such as sellers competing in online markets; autonomous systems exchanging data packages; and people interacting over social networks. Her CAREER project seeks to overcome challenges related to the large-scale and dynamic nature of these systems by developing a theoretical framework that can tractably and robustly capture heterogeneous interactions in large network systems via the use of graph limits.
• Christina Lee Yu, operations research and information engineering (Cornell Engineering), will develop new methods for optimal experimental design and causal inference that address interference – complex dependencies across individuals and time – arising from network interactions and time dynamics. While randomized experiments are widely used to estimate causal effects of proposed treatments in a range of domains – including engineering, health care and tech – many modern systems involve complex dependencies that violate critical assumptions needed for standard techniques to work.