Personal tools
You are here: Home Alumni Grigori Fursin

Grigori Fursin


Grigori Fursin graduated from Moscow Institute of Physics and Technology (Russia) in 1999 and soon after joined School of Informatics at the University of Edinburgh as a PhD student advised by Prof. Michael O'Boyle and researching on how to automate and speed up program optimization using various iterative empirical techniques. Since 2000, Grigori has also been working as an RA on several EU projects related to program optimization.

Grigori obtained his Ph.D. in Computer Science in 2004, spent 3 years at INRIA Futurs (France) as a postdoctoral researcher and in 2007 received a tenured research position at INRIA Saclay. Since 2010, Grigori took personal sabbatical to help establish new Exascale Research Center in France (Intel Labs Europe, CEA, GENCI and UVSQ) as the head of the code characterization and optimization group.

Grigori's long-term objective is to develop self-tuning adaptive computing systems that can automatically adjust their parameters and behavior to different environments and constraints based on empirical, statistical and machine learning techniques. Grigori was one of the leading developers of the world's first machine learning enabled research compiler (MILEPOST GCC) and a founder of the public collaborative R&D center ( to enable collective program and architecture characterization and optimization. Many of his techniques are now actively used in industry and academia.

Grigori's current research page:

Document Actions