I am a Ph.D. student in Computer Science at Texas A&M University, advised by Prof. Nitesh Saxena in the SPIES Lab.

My research focuses on agentic security in high-stakes domains — attacks against and defenses for LLM agents across the harness, agentic loop, tools, skills, and memory, including prompt injection, indirect prompt injection, prompt leaking, jailbreak, and memory poisoning. My current work uses financial trading agents as a representative case (SoK: Trading Agents or Market Crashers?).

Before Texas A&M, I earned my M.S. in Electrical and Computer Engineering from Carnegie Mellon University, where I worked in CyLab on TypeScript taint analysis and vulnerability discovery across the npm ecosystem. I completed my B.A. in Computer Science with a Minor in Statistics at the University of Minnesota, Twin Cities.

Background

Alongside my security research, I specialize in options: pricing models, strategy analysis, 0DTE options arbitrage, and automated quantitative trading. I am a Certified Futures and Options Analyst (CFOA), having earned a perfect score on the exam and received special congratulations from the organizers.

I build and run a fully automated trading agent designed around robustness and security from the ground up. Running on a real-time Benzinga news feed, it has repeatedly entered positions about 15 minutes before the news reached mainstream outlets and profited as the market caught up. The agent is private, but I am happy to share the ideas behind it: in historical backtests, asymmetric spread strategies that keep losses small and let gains run have generated excess returns. I have also built a four-leg options strategy designed to profit from event volatility and the implied volatility (IV) crush that follows.

As Warren Buffett put it, “you only find out who is swimming naked when the tide goes out.” In a bull market everyone looks like a genius; lasting profits come only from managing risk through bear markets and extreme events. That is why risk control comes first and profit second, and why closing a position is always better than hedging it. I hold LLM agents to the same standard: robustness and security matter most when conditions turn adversarial.

The market is a random walk; controlling risk and staying alive is your only alpha.