I am the Computing & Climate Impact Fellow at the MIT Climate & Sustainability Consortium (MCSC) and a postdoc with Christina Delimitrou at the MIT Computer Science & Artificial Intelligence Laboratory (CSAIL). My CV is available here.

I design AI systems that are efficient, adaptive, and robust under resource constraints. My research targets the core systems challenges at the intersection of computer systems, energy systems, and environmental sustainability. I demonstrate that the status quo approaches to designing and operating computer systems cannot respond to the emerging dynamics of rapidly growing AI demand, saturating energy-efficiency gains, and power availability constraints. To address this, I develop new abstractions, system architectures, metrics, and operational paradigms that enable grid-interactive AI systems and raise sustainability as a first-class principle in their design and operation.

Impact: I have made foundational contributions to the design of sustainable AI infrastructure through systems such as EcoVisor, CarbonContainers, CarbonScaler, and WattScope. One of my solutions for dynamic overcommitment is now deployed in all Google datacenters, improving the infrastructure efficiency of global-scale applications like Search, Gmail, and YouTube.

Before MIT: I completed my Ph.D. in Computer Engineering at UMass Amherst in 2022 with David Irwin in the Sustainable Computing Lab, and worked as a postdoc with Prashant Shenoy. During my Ph.D., I was a Student Researcher at Google, working on Borg’s resource scheduler, mentored by Nan Deng and Krzysztof Rządca, and at VMware Research on benchmarking for sustainable computing. I hold an MS in Energy Systems Engineering from NUST Pakistan and a BS in Electrical (Power) Engineering from UET Lahore.

I am deeply committed to academic service. I co-founded the ACM SIGEnergy Graduate Student Seminar and co-chaired the Workshop on Societal Decarbonization (SoDec). I was recognized as a Top Reviewer at ACM e-Energy 2023.

My research has been featured in outlets including WIRED, CBS News, MIT News, Associated Press, Smithsonian Magazine, and the Machines That Fail Us podcast.

If you’re interested in my work on the environmental impact of AI systems, please contact Molly Chase, Communications Manager at MIT Climate & Sustainability, at https://impactclimate.mit.edu/people/molly-chase/.