I am an AI Scientist specializing in building trustworthy AI systems. My work focuses on making large language models safer and more reliable — identifying hidden behaviors, detecting trojans, and systematically evaluating model risks before they become business problems.

At Ericsson's Global AI Accelerator, I built end-to-end AI training pipelines and delivered NLP solutions that outperformed baselines. My research has been funded by NSF, and US Department of Education, which were published at venues including AIware, MSR, ASPLOS, USENIX Security, and SeT LLM at ICLR.

I currently work at the Productivity and Security in Software Systems Lab at the University of Houston, with a postdoc background from Texas A&M University's SPIES Lab and a PhD in Computer Science from the University of Houston.

Visit my projects and publications. Click on the menu button on the top-left to learn more about my experiences.