Octavian Suciu


I am a Postdoctoral Researcher in the Maryland Cybersecurity Center at UMIACS.

I received my PhD in Computer Science from UMD in 2021, working with Prof. Tudor Dumitras. My dissertation focused on Data-Driven Techniques For Vulnerability Assessments.

My research interests include building robust systems for security, large scale measurements and the security of machine learning. I also enjoy collecting unstructured data at scale and analyzing it to discover new cyberthreat indicators.

I contribute to the development of security systems that are used by thousands of practitioners daily:

  • Exploit Prediction Scoring System (EPSS) - a machine learning scoring system capturing the risk of attacks in the wild against vulnerabilities.
  • Expected Exploitability - a deep learning-based system that continuously monitors and collects Web information about vulnerabilities, predicting their ease of exploitation.
  • CVSS Predictor - a large language model fine-tuned to compute the CVSSv3 score for a vulnerability from its description.

My work has been featured in the media, by SC Magazine, Dark Reading, The Register and MIT Technology Review.



Publications (Scholar)






Education




Contact


osuciu at umd dot edu


5104 Brendan Iribe Center,
Department of Computer Science,
University of Maryland, College Park.
MD 20742