Hi! 👋🏼

I’m Sanjana Nambiar, an AI security researcher interested in making LLMs and generative AI systems more reliable, robust, and safe. I’m especially excited about adversarial behavior, trustworthy AI, and uncovering where (and why!) language models fail. Beyond security, I’m curious about what makes LLMs tick; from mechanistic interpretability and reasoning to the reliability of agentic AI systems.

I’m currently a Junior Research Scientist at the Center for Interacting Urban Networks (CITIES) Research Institute at NYU Abu Dhabi, working with Prof. Azza Abouzied and Prof. Christina Pöpper.

News

• Sept 2026: 🎉 Our paper, Data-Void Exploits in Web-Searching LLMs, was accepted to NDSS 2027!
• Sept 2026: Serving as an Artifact Evaluator for IEEE S&P 2027.
• Aug 2026: Organizing and co-hosting the weekly Cybersecurity Seminar Series for AY 2026–2027.
• Jun 2026: Serving on the Program Committee for the AI Security track at ACM ASIACCS 2027.
• Feb 2026: Joined CITIES Research Institute at NYU Abu Dhabi as a Junior Research Scientist.
• Jan 2026: Completed the Technical AI Safety Course by BlueDot Impact, exploring frontier AI risks and safety evaluations.
• Aug 2025: Joined Trevex as an AI Product Researcher, building GenAI-powered agentic workflows and conversational AI systems.
• Jun 2025: 🎉 Published and presented my first first-author paper, JailFact-Bench: A Comprehensive Analysis of Jailbreak Attacks vs. Hallucinations in LLMs, at SiMLA 2025 (co-located with ACNS 2025) in Munich, Germany! [Slides]
• May 2025: 🎓 Graduated from NYU Abu Dhabi with a B.S. in Computer Science and minors in Applied Mathematics and Engineering.
• Apr 2025: Co-authored Style Over Substance: Failure Modes of LLM Judges in Alignment Benchmarking, presented at ICLR 2025.

For more, check out my Publications, Projects, and CV!