* ITBench: Evaluating AI Agents across Diverse Real-World IT Automation Tasks, ICML 2025
* Exploring Large Language Models for Low-Resource IT Information Extraction, ICDM 2023
* Improving Model Performance Using Metric-Guided Data Selection Framework, ICDM 2023
- Patents: Not specifically listed
- Projects: Involved in significant projects such as the ROMEO project, PASIR predictive analytics system, etc.
Research Experience
- Since 2018, using various forms of machine learning to process incident tickets in IT management.
- In 2025, started improving database engine performance using generative models to create SQL test queries.
- From 2018 through 2023, part of the ROMEO project focusing on designing and building cloud infrastructure for data collection, storage, and distribution to processing modules.
- From 2015 until 2018, used the PASIR predictive analytics system to support IBM Managed Security Services' security appliances.
- From 2011 through 2015, worked with the Services Innovation Lab (SIL) developing methods to quickly discover and classify network appliances in data centers.
- From 2008 through 2011, developed ways to locate and track business applications spanning multiple servers in the IBM environment.
- Until 2008, worked on gathering health data from wireless devices.
Education
Ph.D., supervised by Dr. Refik Molva, research conducted at Eurécom, formally enrolled at EPFL in Switzerland.
Background
Research Interests: AI for IT Operations, Database Engine Performance Optimization, Device Discovery in Wireless Networks, etc. Professional Field: Machine Learning, IT Management, Cloud Service Support.
Miscellany
Personal interests include DAB (Eureka-147), device discovery in wireless networks, programming for Symbian and MIDP, and Bluetooth technology (how it works and via the JSR-082 Java interface).