George Candea

EPFL IC IINFCOM DSLAB
INN 330 (Bâtiment INN)
Station 14
1015 Lausanne

George Candea heads the Dependable Systems Lab, where he develops techniques and abstractions for building efficient and trustworthy computer systems. His current work focuses on making performance and energy programmable, i.e., providing first-class abstractions to capture a system’s performance and energy behavior (as interfaces do for functionality) and to use them dependably to design, code, and operate efficient systems.

In the recent past, George also served as CEO and later Chief Scientist of Cyberhaven, a cybersecurity company he co-founded in 2016 with four of his PhD students. Earlier, he was CTO and then Chief Scientist of Aster Data Systems (now Teradata Vantage), a big-data company he co-founded in 2005 with two Stanford colleagues. Before that, he held positions at Oracle, Microsoft Research, and IBM Research.

George is an ACM Fellow (2025) and recipient of the Mark Weiser Award (2024), the inaugural Eurosys Jochen Liedtke Young Researcher Award (2014), the MIT TR35 Young Innovators award (2005), and several Best Paper awards. He holds a Ph.D. in Computer Science (2005) from Stanford University and a B.S. (1997) and M.Eng. (1998) in Electrical Engineering and Computer Science from MIT.

Teaching & PhD

PhD Students

Can Cebeci, Fan Chung, Jiacheng Ma, Roshan Ramachandran Nair, Alexandre Piveteau

Past EPFL PhD Students

Silviu Andrica (2014), Stefan Bucur (2015), Vitaly Chipounov (2014), Rishabh Iyer (2023), Horatiu Jula (2011), Baris Can Cengiz Kasikci (2015), Volodymyr Kuznetsov (2016), Solal Pirelli (2024), Jonas Wagner (2017), Cristian Zamfir (2013)

Past EPFL PhD Students as codirector

Radu Banabic (2015), Arseniy Zaostrovnykh (2020)

Courses

Principles of computer systems

CS-522

This advanced graduate course teaches the key design principles underlying successful computer and communication systems, and shows how to solve real problems with ideas, techniques, and algorithms from operating systems, networks, databases, programming languages, and computer architecture.

The software enterprise - from ideas to products

CS-311

The course teaches the journey that software teams take from nascent ideas to products that solve real problems for real people, in an era where AI agents perform much of the implementation, and the engineer's craft increasingly centers on specifying, directing, reviewing, and validating their work.