I’m Juliette Pluto, a Staff Software Engineer at Google DeepMind in New York City. I work on adversarial robustness for frontier AI systems; lately, mostly prompt injection: what an agent should do when the data it reads tries to give it orders.
I started DeepMind’s first effort to defend against prompt injection, and now lead our prompt injection evaluations. My work spans training Gemini, improving adaptive attackers, and practical defenses for production systems.
Before that, I worked on trusted execution environments and privacy-preserving ML at Google in London, and helped scale N26’s web platform in Berlin.
Read my CV for the details, say hi at [email protected], or find me on GitHub and Twitter.