Pantazis‘s research involves designing and developing novel tools and techniques for programming, analysing and testing complex asynchronous software, ranging from device drivers to distributed systems.
Pantazis is currently working on two projects for his PhD: P#, a new event-driven asynchronous programming language that is co-designed with static data race analysis and systematic testing; and Whoop, a fully automated tool that statically analyses Linux device drivers for data races, and then exploits race-freedom guarantees to accelerate bug-finding.
His industry work experience includes internships at Microsoft Research and Codeplay. Pantazis’s research is generously funded by a gift from a href="http://www.intel.com/content/www/us/en/homepage.html" target="blank"Intel.
Pantazis’s research interests include automatic analysis for detecting data races in concurrent Linux device drivers, and the P# language for reliable asynchronous programming.
Uncovering Bugs in Distributed Storage Systems during Testing (not in Production!)
14th USENIX Conference on File and Storage Technologies (FAST'16)
Implementing and Evaluating Candidate-Based Invariant Generation
Fast and Precise Symbolic Analysis of Concurrency Bugs in Device Drivers
30th IEEE/ACM International Conference on Automated Software Engineering (ASE'15)
Asynchronous Programming, Analysis and Testing with State Machines
36th Annual ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'15)
Automatic Verification of Data Race Freedom in Device Drivers
2014 Imperial College Computing Student Workshop (ICCSW'14)
Engineering a Static Verification Tool for GPU Kernels
26th International Conference on Computer Aided Verification (CAV'14)