Avatar

Ravikumar Balakrishnan

Principal ML Engineer

AI Software & Platform

Ravi Balakrishnan is a Principal ML Engineer at Cisco where he leads the efforts in multimodal security of modern GenAI systems. In addition, he is actively involved in several efforts in securing and red teaming agentic systems. Prior to his role at Cisco, he has been a seasoned industry researcher for 11 years across HiddenLayer and Intel Labs with a diverse portfolio of research areas spanning privacy-preserving, adversarial machine learning, algorithms for wireless communication systems with 25+ peer-reviewed publications and 50+ patents including a best paper award at MLSys 2023. In his past roles, he has served as a liaison across industry-funded university research partnerships, including collaborations with NSF, SRC, and academic institutions, with responsibility for identifying, translating, and integrating promising research into product innovation. He actively serves in top conferences organizing numerous workshops and panels on a broad range of machine learning and security related topics. Ravi holds a PhD from Georgia Tech from 2015.

Articles

Reading Between the Pixels: Assessing Prompt Injection Attack Success in Images

6 min read

This post is Part 1 of a two-part series on multimodal typographic attacks. This blog was written in collaboration between Ravi Balakrishnan, Amy Chang, Sanket Mendapara, and Ankit Garg. Modern generative AI models and agents increasingly treat...