About

Background

Originally from Houston, TX, I'm an AI research scientist working at the intersection of large language models and adversarial security. I graduated from Northeastern University with a B.S. in Mathematics and Data Science in 2023. At Booz Allen Hamilton, I've designed defenses against LLM jailbreaks and suffix attacks, studied agentic policy-following under adversarial pressure, and built state-of-the-art systems for binary analysis used in the US Intelligence Community. I'm currently a PhD student in Computer Science at UMBC, where my research focuses on the security implications of increasingly capable AI systems — both what they can be made to do, and what they can be used to defend against.

Research Interests

• Agent Red-Teaming & Policy-Following under Adversarial Pressure
• Binary Function Similarity and Information Retrieval
• Representation Learning
• Adversarial Machine Learning

Publications

• [1*] "Binary Function Similarity has an Edit Distance Problem". Under Review, SIGIR '26
• [2*] "Binary Function Retrieval and Distraction". Under Review, IEEE WorMA '26
• [3] "Is Function Similarity Over-Engineered? Building a Benchmark". NeurIPS 2024.
• [4*] "How to Protect LLMs from Jailbreaking Attacks". Technical Whitepaper. 2024.

* First-author

Contact

Want to chat? Reach out on LinkedIn!