Sai Praneeth Karimireddy
Assistant Professor, Thomas Lord Department of Computer Science (& by courtesy, ECE), USC.
I am an Assistant Professor in USC's Thomas Lord Department of Computer Science, and by courtesy in the Ming Hsieh Department of Electrical & Computer Engineering. Before USC, I was an SNSF postdoc working with Mike I. Jordan at UC Berkeley, and I completed my PhD at EPFL advised by Martin Jaggi. My work has been deployed in industry at Meta, Google, OpenAI, and Owkin.
See the group page for how to work with me.
Research
AI is already changing our world in unprecedented ways. Questions that were once mostly philosophical — what is truth, is there a difference between observable behavior and innate traits, what improves human welfare — are now becoming imperative. I take a principled approach to carefully formalize, understand, and answer such questions. It is an exciting time for research :)
Topics I am currently thinking about:
Trustworthy AI. What does it mean for LLMs to be safe or private? How can we evaluate and characterize LLM capabilities and behaviors?
AI ecosystems. How will AI agents interact with humans, society, and each other — and how can we understand and shape these emergent phenomena?
AI in health and finance. Can we translate these insights to deploy AI in high-stakes settings such as healthcare and finance?
Some selected papers:
What are the fundamental limits of black-box auditing?
How do you feed context to an LLM in a differentially-private manner?
How do you model multi-agent cooperation and design incentives without monetary rewards?
What distinguishes important corner-case data from noisy outliers?
Awards
- Amazon Center on Secure & Trusted ML Award
- Capital One Fellowship
- SNSF Mobility Fellowship
- Patrick Denantes Memorial Prize — best PhD thesis in computer science, EPFL
- Dimitris N. Chorafas Foundation Prize — exceptional applied research