Hello, friends!

I research AI safety, focusing on whether AI supports or undermines human flourishing. I co-architected HumaneBench.ai, finding that 67% of leading models can be manipulated into giving harmful advice through adversarial prompting. I’ve built evaluation infrastructure that organizations actually adopt—from HITRUST certification unlocking millions in revenue to AI-powered cybersecurity to HPC benchmarking and optimization serving 75M CPU hours annually.

Currently, I support large-scale computational research at Georgia State University, helping research teams deploy responsible AI infrastructure. I also organize hackathons and community events advancing humane AI development.

Previously, I developed RL-powered solar microgrids at Microsoft Research and built AI-driven network security at OPEX Systems.

Explore my Research, my Resume, and my Blog.

News

My work has been featured in:

• Built In: Are Chatbots Acting in Our Best Interest? A New Benchmark Says Maybe Not.
• TechCrunch: A New AI Benchmark Tests Whether Chatbots Protect Human Wellbeing
• Microsoft Research Blog: AI-Powered Microgrids Facilitate Energy Resilience and Equity in Regional Communities