 I'm working mostly these days on form verification, which is trying to get programmers to write code that is fully correct, has no bugs. We're also working a little bit on fault tolerance. In general, the goal is build systems that are always available and cannot be brought down by other hardware failures or software bugs. Most of these days, we're trying to tackle large, complex distributed systems. We want to be able to have a methodology so that people can write a complex distributed system, can implement it all the way down to, let's say, C code and still be able to have that code be absolutely free of bugs. Our main goal here is reliability. Make sure that the systems, you know, if you build a system, it will actually remain available. They will not just be correct, you know, once. They will always remain correct. They will tolerate, you know, they won't have bugs and they will tolerate any machines failing, any parts, some parts of your system going down. Your system will still be available just like you're used to, you know, your Facebook page always being available. In order to do that, what's really, the key to all that is to try to understand the system is better. So there's all these little nuggets of discovery that we come up with and this is kind of these aha moments where like my student comes in and say, you know what? Remember that question we had? I now know how to solve that. What if we did that? And you're like, oh yeah, that looks really good. And then we move on to the next problem. We envision a world where it would be unthinkable for someone, you know, a major company like Google or Microsoft or Amazon to deploy a piece of code that will not be formally verified. So it would be part of the normal cycle of a program that you develop it, you formally verify it and then you deploy it. You never deploy it just by having tested that program because, well, you may be missing some of the bugs by just testing. So the grad students start off as students if you want, that's my philosophy that they should not remain students for long. They need to become independent researchers and that's the whole point of them acquiring all the knowledge that they need, the expertise so that a few years down the line they won't need me. They will be able to do their research on their own and I'm trying to help them with my experience so that they will become soon independent researchers.