LUCASVADLAMUDIVADLAMUDI/vud-luh-moo-dee/

I'm an AI engineer focused on security, and a junior at Texas A&M studying computer science with minors in cybersecurity and artificial intelligence. At the Texas A&M Global Cybersecurity Research Institute, I lead research on AI control backed by $35K in funding, studying how continual learning wears down the monitors that oversee AI agents, and how to fix it.

My previous work focused on using AI agents to enhance security operations centers, automating alert triage and enriching investigations with threat intelligence. I also benchmark inference systems for mixture-of-experts models, optimizing hardware and serving configurations to maximize intelligence per wattAn efficiency metric that decides the useful model capability per watt of power drawn on edge hardware..

I'm co-founding the Texas A&M AI Safety Club, bringing conversations and projects focused on reducing the risks of advanced AI to Texas A&M University while giving motivated students practical opportunities, mentorship, and clearer pathways into the AI safety field.