Krishnamurthy Dvijotham

Krishnamurthy Dvijotham (Dj) is a research scientist with a focus on developing safe, reliable, and secure AI systems. His current role as a Research Lead at ServiceNow Research (2024-2025) builds on his extensive experience at Google DeepMind (Researcher, 2017-2024), Pacific Northwest National Laboratory (Researcher, 2016-2017), and a postdoctoral fellowship at Caltech's Center for Mathematics of Information.

• Research Interests: Mathematical optimization, control theory, AI robustness, differential privacy, neural network verification, power systems optimization.
• Scientific Awards: Best Paper at ICML 2024, Best Paper at UAI 2018, Best Paper at Constraints 2016, Best Student Paper at UAI 2014.

Dvijotham's work emphasizes the application of rigorous mathematical frameworks to AI systems. His publications span certified robustness, adversarial defense mechanisms, privacy-preserving learning, and power grid optimization, reflecting interdisciplinary contributions to both theoretical and applied domains. He has mentored numerous PhD students and postdoctoral fellows, many of whom now hold academic or research positions at institutions like UC Berkeley, Google DeepMind, and Microsoft Research. His recent articles highlight advancements in diffusion models, selective deferral systems, and formal verification techniques, with a strong focus on security and reliability in AI deployment.