luke farchione_

phd student · security researcher · bug bounty hunter

i'm a cybersecurity PhD student at UCCS in the Xu Lab, specializing in large language models in cybersecurity. when i'm not training models, i'm breaking things.

latest

from one tracking URL to every DoorDash customer's PIIP1 · still exploitable · june 2026reading anyone's private Meta AI conversations$5,000 bounty · march 2026

research

  • CASSANDRA — new state-of-the-art model for CTI NLP and automated TTP extraction. in progress
  • Aegis — cybersecurity visualization tool with an 18-attribute competency framework. in progress
  • CVE-LLM — real-time vulnerability querying using NIST NVD, MITRE, and Exploit-DB with vector retrieval
  • Bili — open-source framework for evaluating LLMs across cybersecurity domains. NSF funded

elsewhere

githubemailresume