On GraphLens, I developed a prototype for discovering and managing organizational assets, with a human-in-the-loop process to improve the completeness and reliability of its asset graph. I also integrated an AI agent that proposes graph updates and asks analysts clarifying questions.
For ThreatLens, I developed a framework to prioritize vulnerabilities using public evidence, achieving 80.0% Recall@20 and 95.9% Recall@50. I co-authored a paper on the findings that was accepted for a poster presentation at ACM CIKM 2026.
