AI and machine learning researcher, and a Computer Science PhD student at the University of Texas Rio Grande Valley — in the Structures & Computational Intelligence Lab. My research runs from AI safety and mechanistic interpretability to LLM evaluation and trustworthy ML for real-world infrastructure, building on my M.S. in Computer Science at Texas Tech University.
25 years old
Lubbock, Texas, USA
I'm Rahman, an AI and machine learning researcher who cares about systems that stay trustworthy once they leave the lab. Most of my work lives in AI safety — LLM security, mechanistic interpretability, and adversarial ML — including SAEGuardBench (jailbreak detection with sparse autoencoders), MergeSafe (backdoors that survive model merging), and CompToolBench (compound tool-use evaluation). I'm a Computer Science PhD student at UTRGV, in the Structures & Computational Intelligence Lab, where I bring that ML background to intelligent, resilient infrastructure — making foundation models reliable and grounded, and using machine learning for infrastructure sensing and assessment. Before this, I was a Graduate Research Assistant at Texas Tech's TransTech Lab, building real-time roadside sensing for a TxDOT pedestrian-safety project, and I earned my M.S. in Computer Science at Texas Tech. If you're working on trustworthy ML, AI safety, or AI for real-world systems, I'd love to compare notes.