 So, Ryan, thanks so much for taking our little tour of Amsterdam and talking a little bit about what you're working on and what you're doing. And do you want to go ahead and introduce yourself a little bit? Sure. So I'm Ryan Hallacy. I'm an engineer at NVIDIA. I work on building the infrastructure for GeForce Now. GeForce Now is a cloud gaming service. So kind of like what I said, people, if you ever wanted to get access to a nice 3080 and stream cyberpunk on it from your phone, you can use GeForce Now. Nice. Nice. And is it available? Like you can go out and pay for it now? Yes, you can. Sweet. I may have to check it out more. Like I remember reading about it, but I had never actually kind of gone and looked at it. And because, you know, like I have kids and a job and so playing a lot of video games is not high on my list of things I can do anymore, which is depressing. But yeah, so we, I know you just gave a talk. What was your talk about? So my talk was about my experience as an end user of Cuba. So GeForce Now, the infrastructure behind it heavily leverages Kubernetes and Cuba. And so we went over kind of silly use cases that we have and how we use Cuba and how we use Cuba and how we like it. Okay, cool. So do you code every day or are you management now? No, I kind of do, let's see, I do a lot of coding when I can and do some architecting. Yeah, I was talking to somebody else earlier. It's like the only downside about being in the software world is a lot of us got into it because we like to code. When you get more senior, you get to code less and less. You know, but it's nice to be able to keep your hands dirty sometimes. Yeah, I'm involved heavily in the community. Like I'm one of the maintainers of Kuvert. So I'm active in the community and contributing features, design proposals, stuff like that. And then very active in a few of the six in the community as well. So I feel like I know this answer a little bit. So what brought you to kind of open source and or Kubernetes in the first place? Well, there's a lot of things. I mean, I like Kubernetes. So me personally. Yeah, you personally personally. So for me, it's a really project is really exciting. I mean, it's like it's really cool because I like understand it. I sort of connect to it at a very low level as someone who like understands infrastructure and likes to build things and understand like what it takes to launch a workload on my local laptop. Yeah. And then think about like all of like the bash scripts that I ever had to deal with in order to launch that workload. And then I think about like why the Kubernetes does it. It's sort of like encapsulating a lot of the knowledge that people would go through. I know those experiences that you go through in order to actually launch a workload became like kind of a lot of way I look at Kubernetes. I mean, you can think about like deployment and stuff, right? I could I could do a create a deployment like resource if I wanted to. If I had like a bash job that's constantly running and it's constantly replacing things. If I noticed, you know, watching events, something going down on the bus, like you could almost do this. And it's like, and so I kind of think of it like that. And so to me, as someone who's really curious and interested in building new and exciting things, it makes a lot of sense. And someone who's had a lot of experience in this area. So that was something that was really exciting to me. And I think that's why a lot of people are excited about it. They really connect to that part of Kubernetes. And that's really why the community is so exciting and there's a lot of people here. And so it really kind of feeds the energy and why it's such an exciting project. And that's why I like being involved in the community. Yeah. Yeah. And kind of related. So what brought you to open source in general to begin with? Sure. For me, I always found it fascinating. I had even before, you know, I contribute to my first open source project. I always thought it was kind of interesting to share my projects with people. And so I was well before I understood open source and really would have met. But it was something that to me that I found it interesting when I had my friends and I had work that I wanted to show with them so they could use it. And what I didn't really realize is I was actually doing open source. I was building my own community of software. And so naturally, like when, you know, hearing open source and what like there's a whole world of open source eventually when I, you know, graduate school. And I was like, well, this seems like something that I already do. And it seems like something I already like. So I'm really interested in this. I kind of wanted to give this a try. That's cool. That's cool. And so, so you were, and that was because I know you were a Red Hat. And so that's what kind of brought you to Red Hat. Yeah. I had heard of Red Hat and I had been familiar with Linux and familiar with Red Hat software. And so for me, Red Hat was was like, like a dream. Like it was like, wow, this is the Linux place. Right. I definitely wanted to give this a try and work. And this is the place center of open source. So it was very appealing to me. Yeah. Yeah. That's really cool. I think it's funny, right? There's, you know, at Red Hat, right? There's people who definitely are coming in kind of with a leg up right where you actually know what Linux is, right? And have been using it or really interested in it. But then there's other engineers, right? Who like have very little to no experience with Linux. And they're like, you know, they know they've heard of Red Hat, you know, but it's just kind of interesting. But then, you know, the, I would say a large majority though are coming in with, you know, like because they're passionate about, you know, open source and Linux and, you know, stuff like that. And it's really, you know, quite cool to be able to do that sort of thing. Yeah. I see that a lot. Red Hat, in my experience at Red Hat, I saw that a lot. That was the thing you see is this passion for open source. It's very, very prevalent in the culture and how, and at the company. I made a bit of a wrong turn. So we're doing a different tour of Amsterdam than I've done before in error because I got forced into a left right turn. I didn't mean to get forced into it. So that's why you're getting a little bit of a different tour. Yeah, exactly. So hopefully we'll make it back to the correct route eventually. But that is why we have the, you know, Nav going. Because I don't know Amsterdam, but speed bump. So, yeah, so sorry, back to our regular schedule and programming. So where does kind of Cuber come into play? Like you said, it's kind of involved in GeForce Now. But what, like how does it kind of do that? Sure. Or what does it do there? Yeah, sure. So what I tell people is like inviola and invidians specifically to GeForce Now like virtual machines. And it's, you know, virtual machines are a very popular solution and we use them a lot. And so this idea of Kubernetes is out there as an ecosystem in a way of doing things. And microservice architecture using containers. And it's sort of, and there's all this big community around it. And so, you know, we like VM. So, well, okay, let's try, you know, can we, can we go to this kind of model? Like, can we actually look at using, you know, a microservice approach in containers? So this was kind of our thought process. And so when looking at what our existing solution was, we still want to maintain the existing investment and using virtual machines. And so with Cuber, Cuber provided the ability for us to actually go to Kubernetes and get the leverage to the ecosystem as well as continue to be able to run virtual machines. Right. And it's a natural fit for us to be able to continue to with that investment and get the best of, because we like Kubernetes. We kind of want to use some of the things around that Kubernetes has. And so we do that for some things and services that we have. We also like VM. So we use Cuber to continue to run VMs. Gotcha. And so, you know, I'm sure, you know, the infrastructure isn't just like those VMs, right? So is there, like, do you get to take advantage of kind of the rest of Kubernetes? So in other words, like, do you have containerized, you know, components of the system that like integrate with those pieces? Yeah, we do. So, so we use with Kubernetes, like we integrated, you know, all the layers, C&I layers and CDI layers, all everything we in the CRI layer, everything. So we were integrated all those layers. And then also we do have our own components, like a lot of the stuff that came along, you know, after we like, hey, we want to use this, you know, we needed to come up with components to run as pods. So work with our existing stack. And so a lot of, all, pretty much all of that work was done as containers. Okay. Very little of it. Once we started using Kubernetes became, like, you know, we don't want to continue to use VMs for our management or control plane. We want to try and get as much onto containers as possible and leverage Kubernetes. Right. So for a lot of the new stuff, the new stuff, I should say, is the, is the, is using containers. And so some of the stuff, even we've, you know, anything that we had to use in the prior generation architecture, we even try to move to containerized workload. In some cases it works well, some cases it doesn't. It's just, it just is what it is. But so it, but it works for us, like if we continue to keep VMs around, it's fine. But in cases that, you know, we find it fits, then we use it. Right. We use containers. And that's, that's what I'm, what I think is kind of most interesting of, like, kind of talking to the user of the, of the tool chain is like, and, and you've gotten that to work. Right. So, like, you can have, you know, those operating together and it seems to, you know, consistently do what you wanted to do in a sense. Yeah. Yeah, exactly. Yeah. We've been able to get what we want out of it. The thing that we like always say we love, really love about Kubernetes is like, that, like I was talking about earlier, like being able to revive workloads, sort of have that ability to, for workloads to return when they fail. Something that like their resiliency, right? Yeah. That the platform offers. We really like that. And then in the community as well, like the, the feature velocity and then, and also the ability to contribute, like to the, the existing velocity is also really exciting too. Yeah. And we like turning back. Right. And so do you, do you feel like, you know, are, are you regularly contributing, you know, the features that you think are important and that you need back into the upstream? Yeah. Yeah. So we have, we have influence in different parts of the community, some in a little bit of Kubernetes and then some more in Kuber. And so like we've, we've had some very specific ones like with Kuber. We've done VSOC. We collaborated with Google to create VSOC and contribute back. We've created some new APIs in Kuber. We actually go in, we call it virtual machine pools. It's like an AWS. It's like auto scaling groups. Right. And we, so we created that because it was perfect for how we represents how we use our workflow. We, we have lots of VMs and they kind of, they come in and out constantly have churn. And so we kind of used the, like you use like the golden image VM kind of model where you have like one VM that's like stock and then you spin them out when you need them. Yeah. Like we, we do like we basically have like images that we distribute throughout our zones, like whenever we have updates or changes. Yeah. Yeah. And so we, we do do that. And yeah. And so yeah, we've been, we've contributed to the community a bunch of different ways. Those are just some, some examples. And we've got some other things, like even in the Kubernetes community with some colleagues that have been working in different areas for, and resource allocation and some other changes for other ways that you can make GPUs available in Kubernetes. Yeah. Yeah. So, you know, I kind of been asking this from kind of a bunch of different perspectives. But like, so why do you think, like, do you think that Kubernetes, you know, if it wasn't kind of an open source project, like what, you know, what's the, what's the trade off? Like what might be better or worse about it that you think, you know, is a benefit to being open source if there is one. Sure. Yeah. I haven't really thought about it, but, you know, being, being open source, I mean, I think, I think what it, what it does is it, I mean, it gets like I was saying that the velocity, I think the feature velocity, the developer velocity, there's just a lot of access to different ideas and use cases. And what we found is like when, when we look at the project, we see, we see like, like 90% of a solution. Like we get like, we get a lot of things, right. And that's great. Like that's actually probably around what we want. And then like, we want to do the rest. So like, if it was maybe closed source, I mean, I don't know. I mean, it could vary like and how we, how we could want to approach it. But we kind of like, like, I guess the way to look at it is like, we kind of like doing the management infrastructure. Like we like that. And then we want to be responsible for, we don't want to like go and like, you know, we sell GPUs and then go rent them. From someone else. We want to actually, we actually want to put GPUs in our data centers and run them in the world class way. And that's, that's what we do. And that's what requires us learning new things and trying to make it work. So an open source solution is a perfect fit for us. Right. Right. Yeah. Yeah. I was talking to somebody earlier that made a really interesting point was that, you know, you know, when you, when you look at software, right, any software makes a whole bunch of choices that are like tradeoffs between various things. And, you know, it's just kind of required, right? And the, the, you have certain biases around where those tradeoff should be, depending on the use case and all this different stuff. But he was saying what was really interesting for him about open source was that he can look at the application that he is going to build on top of essentially and understand where their tradeoffs and bias choices were so that he can make in his code, either similar or complimentary choices. So that he can get, you know, really good pairing with the open source software that he never be able to do with proprietary software. I don't know. Does that resonate with you? I thought it was super interesting. Yeah, it does. Because so here's a pretty, pretty simple idea. Like if I wanted to find the developer who contributed, I could find the developer. Right. If you had to. Right. I could get blame. Right. Right. I could find this person and I, and I could look it up. And then, and also with, with communities, you see like all the, the conversations are had on mailing lists and on, you know, other sort of the IRC or Slack, whatever. Like it's out in the open. And so you can read all the context as much as you want and to see what the biases were. Right. Yeah. What were the, what was the context that went into this decision? Right. And, and, and you know, and you can have a conversation with the person. And, and, and usually that person, I mean, in a project that's open source is probably still around. Right. It's a future. It's likely the case. Right. And you can talk to them about it and bring your own perspective and see if it makes sense. Right. Incomplain. Incomplain. Yeah. And you can complain. And so, yeah. And then what can result from that? Well, it could be some new code, a new use case, a new direction. Or maybe, you know, maybe it's like, maybe you were wrong. Right. You understand. Maybe you're. Oh, I didn't think of that part. Yeah. Exactly. Maybe because the reason this person did it this way is because they were thinking like you were before and now you just needed the conversation to get more context. Right. And it's a wealth knowledge. Like there's not just, not just like the code itself, but the access to the information is so helpful for making decisions. So one of the other things I think is interesting and, you know, especially coming from, you know, you spend some time at Red Hat, from, you know, how much has open source contributed to your ability to write? So that's a good question. So. And I mean English or, you know, spoken word, not code. Yeah. So that's funny. So I've thought about this before like a, because as someone, I never really liked writing, like it wasn't something that was my favorite thing, but I quickly learned that like you have to be really, really good with, with your communication over different channels and not really, I mean, wouldn't even say spoken word because like most of it's not like it's over text based channels. You have to be really good at getting your point across and communicating exactly what you mean. Or you're going to have a hard time. Like it's really, it's really easy just because people, you know, time is finite and as well as, you know, people's attention. So like, you know, how do you get someone's time to be able to communicate what you need? And also like how you get it so like, you know, make people angry or make it seem like you're not listening to what they're saying. There's some really, there's some nuances that are really important that go into it. I think it's like a skill. Like learning, you can always say like learning to communicate in the open source world is like, is a skill. There's a way to do it. And it's not just like writing, which had my writing has improved. Yeah. Like in order to do it. But there's certain there's a style to it. And that's something that I definitely think my writing has improved over time. Yeah. I know like from working at Red Hat, I know my writing got a lot better, you know, and it's just I find it really interesting that, you know, that's certainly not something they they teach you when you're in college, right? That you're, you know, that your code will that, you know, your computer science degree is going to revolve around a lot of writing. You know, or as you, and when I also found when I got like into management and stuff too, it was like, oh, I also have to be able to read contract law, you know, and, you know, and it's kind of funny how all these other skills are also required. Yeah. What I always say to like junior devs is like, you can communicate through your code and that that's valuable in and of itself, but your code is not enough. Like it's just one tool that you have in order to be successful, because a feature isn't just code. And it's one of the things that you have to teach junior developers. It's like it's much more than that. It's it's you have to you have to explain yourself. You have to explain where you're going. The purpose of this, you know, where what it is that, you know, that you're designed is everything like that. It's so much more than just the code. And you see that sometimes in open source products and this is like one of those gaps that always like that always drives you nuts. Like you see someone show up and they just drop code off. Yeah. It's not it's not going to work. Like what is that? Like how is how someone who's reviewing this and works in the community. How do I feel about this? I don't feel really great about this because, you know, well, I mean, what is what does this mean? Like are you like is this perspective? Like is this right? Like I don't fully understand where you're coming from. Your code means something to me, but I can't read everything that all of your desires from just your code. You need to talk to me about it. And I also understand how you want to maintain this because it's right. There's there's that too. So it's it's more and it's almost like a relationship. You have to build and that's that's part of it. Yeah. One of the things, you know, so I try to convince the students of right is that I'm like, you know, one of the most valuable things you can probably do while you're in college, you know, you have in the grand scheme of things, you have a relatively large amount of free time. Right. And so contributing to some, you know, important open source projects in literally any way will do wonders for your not only kind of your resume, just to kind of show up for, you know, in context, but like, but like your ability to communicate but also like really the big thing is for me, if I'm hiring somebody, seeing that they contributed to an open source project and had their code actually accepted means that they not only can write the code, right, but they can communicate about it. They can, you know, they can, I jokingly referred to it as like they can play well with others provably, you know, and it's such a big deal. Yeah, absolutely. Yeah. Students listen up. Right. Right. Yeah. Yeah. I mean, it's absolutely. I agree. It's there. There is a, there's a skill to it. And it's something that really anyone who understands or is interested in coding can do. It's something that can learn. It's, it just takes a little bit of practice. Right. A little bit of trying it out and, and you can get a, you get a feel for it. Yeah. Yeah. And you know, one of the things I learned that this was even back in consulting is like, you know, make sure you have somebody else read your stuff, you know, often, you know, when it's those important emails or whatever, if you can give in somebody else to read it first and make sure they understood what you wrote, you know, in the way that you intended them to, that can be a big help. Yeah. That's a good, I do that. I read it out loud. I read it on myself. That's the other one I found to be helpful because we hear it. And you're like, oh my God, this is. Right. Right. Right. There's something different about that. I sound insane. Yeah. Totally agree with that. Yeah. I'll, I do that as well. But yeah. So that's totally with you. So we're sorry. They got a little, a little dicey or there with people trying to cross the street. So what other, you know, how much involved are you in other parts of Kubernetes? Is it primarily Qvert? Are you working with other components that you yourself are, are looking at? Sure. So mostly my time has been split into Qvert. And, and then I try to like get involved in other areas, but it's, it's, it's been hard for me too. I've hoping like, since I do a lot of work on Qvert as part of the scalability group, that we can eventually like work with the Kubernetes scalability group. We might do one of the only projects that has an entire significant scale. And I know Kubernetes does. So it's like, I'm kind of thinking like maybe we should, you know, we can meet up and talk more about it. So I was actually at the conference able to meet from the, the one, the main, one of the maintainers of that group. So, and we're going to try and meet up in the future. So I'm hoping that like that's another way we can get involved because I actually think we can learn a lot from each other where Qvert is very focused at scale. It's focused on like our, our goals focused on the control plan and scaling the control plan, like being like a good member of the Kubernetes cluster. It's important. It's really, it's not on. So you need to be a good member of the cluster. Well, it's a different perspective with Kubernetes. You're talking about pods and nodes and how many you can get and how many, how many they perform and so on. But the reality, like the scalability of Kubernetes or Qvert heavily depends on Kubernetes because all this stuff, like once you get to a point where like provisioning storage and network and all this stuff, we have no visibility. Like we're just, we're looking at just Qvert, but we want to know, like people ask me like, okay, how long does it take for me to provision a VM if I'm using sound? I have no idea. But like, like we could maybe find that out. Like if we, maybe if we, you know, took advantage of the way we measure currently and Qvert and becoming some new ideas, maybe there's a way we can kind of, you know, find a way to do this. And so that we can, so I can eventually answer that question and help them too. Right. Yeah. No, that's, that's cool. I mean, and you're also, I mean, you're bringing a, you know, a real use case to scale. Right. And so I think that's also kind of benefit, you know, you can kind of hear other perspectives. It's kind of like everything else with the, you know, everyone has slightly the more you start to see where the similarities are and where the differences are, especially ones where it like, at least for me, a lot of times is the most interesting bits are like where I thought something was obvious. And I was either wrong, like it's just, it wasn't true at all. Or nobody else thinks it's obvious because for whatever reason, the thing I'm working on makes it very obvious. But what most people do, it's not right. And so I've, that's one of the things I find really interesting when you're having kind of those kinds of connections with other people who are kind of using the same tool chain. Yeah, I know the, I know the feeling like sometimes you just different perspectives and sometimes you have different contexts. Right. So really aligning on things. Yeah. All the time you go through that and when you have conversations about features or bugs. Right. Right. So one of the things, you know, the show is called KB Insider. You know, what do you, you know, see in Kuvert or wider Kubernetes, you know, in six months or a year that you're kind of most excited about? What's the, you know, what do you think is going to happen next that will be really cool? Sure. So there's like two specific things that like I'm really interested in. One is, is this, the DRA feature it's called Dynamic Research Allocation. This guy from NVIDIA who's been who's been working on this and pushing in the community. It's really exciting for us because when we do provisioning now with the device plugin, the design of the device plugin and what it's really meant to do is it's meant to serve. I'm going to give, I'm going to provide a GPU for, for a VM or container, but I'm just going to provide it as pass through. I'm not going to be able to also switch dynamically between pass through and VGPU based on the workload that's coming in. And this is actually how we want to run our data centers is we actually want to, because we think about capacity, right? We, we have a physical device pass through or we can be providing as a VGPU. And so it really depends on what we get. So we really need to switch through each type. But we really can't do that with the limitation. So with DRA, this feature, we will be able to actually do this properly, which is really exciting for us because it's going to help us solve a huge number of bugs and race conditions. We currently have to work around this. And also like it's going to help us solve this problem. And it's going to help us solve the problems that we're facing and the things that we're doing before. It's doing it rather than the one that you home grew. Exactly. So that's one of them. The second one is some of the HPC work I see and hearing about. Like I think one of the things like for us that we really care about is being able to do HPC workloads, high performance computing and things like CPU pinning and things like that. And that's where we feel like there's a lot more work that can be done. And so I'm hoping that there's going to be some work and I'm at least from what I'm hearing is that there's some work for some plugins or extensions to be able to actually get down and make more control. So I'm hoping that there's going to be some work. And I'm hoping that there's going to be some work and I'm hoping that there's going to get more control to enable high performance computing loads like maybe like the L3 cache bandwidth control like specific access like, let's look at the CPU and let's break it down into chunks. But let's do it using the Kubernetes API. It's not Bright, reinvent it. Yeah. So that's what we're looking forward to seeing. Yeah. Interestingly, I was talking to a guy from Huawei, Kevin Wang who's involved in the Kuber community. But he's also working on a project called Volcano. know if you heard of this. So you should check it out. But it's it's basically it's kind of bringing you know trying to help solve like scheduling problems for like AI scenarios and having different kind of schedulers but it's kind of related to what you're talking about of kind of saying hey I want to prioritize how things happen differently when I'm doing this kind of workload. And you know so I don't know I think there's a potentially some interesting stuff there for I hadn't heard of the project either I was like oh that's really kind of neat. But then again I have this weird interest in schedulers that I don't know why but I've always been really interested in them and never really gotten like sat down and like worked on them. I just think they're neat. You must be an old school Linux guy. Yeah right right exactly. Yeah yeah I've been I've been a Linux guy for a long time though. I still remember installing like literally having a stack of floppies like this. I don't know if the camera can see me but like you know 25 floppies you know to install oh it's just blanked. You know what I'm talking about the old versions around yeah but no before that or it's like the off the shelf one. VBAT was one of the maintainers for a long time. I can't believe I'm playing on it. It starts at an S but it used to be a super popular distro but it's not so much anymore. But yeah it used to come on CDs in the back of magazines and but in those days it was like you had to like pay attention to the the settings for your monitor because you could literally light it on fire you know if you didn't set the resolution right. Yeah I can't believe I can't think of the name of it. You're the Linux next up and can light your monitor on fire. Right right every year for the last 20. Yeah it's actually one of the big things that's what kind of got me involved with Boston University being with when I was at Red Hat was trying to look at I wanted to know what like student developers are doing today right so that I could try to figure out how could we what could we add to the Fedora desktop to like really enable developers you know it's like Mac you know has been a lot of the developer desktop in recent history but it's at least for me right it's almost by accident right it's that they have a terminal so that's a huge win but then is it you know but but then it just has a you a good UI and it doesn't break right but there is a really is it really bringing that much to the table about developer enablement and like could we bring that to the table in Fedora which I always thought would be really cool you know and just like how much that you know we're I was talking to somebody else too about this new concept of not new but newish platform engineering and so what can we do in platform engineering in the you know in our our laptops to actually enable engineers much better right because I think there's so much opportunity there that we we're so busy trying to make sure that your email works that we don't you know kind of go kind of sideways and like what can we do for like developer tooling you know yeah no that's that's cool I still remember like when the first time that I got access to Linux I don't even know if it was like I think it was Fedora or something I'm not even sure I'm anymore yeah but I just but me oh no and I know what it was it was like it was it was a rel 3 machine in a lab that's what it was it was a rel 3 machine I just thought it was really cool because I just remember the terminal and I remember the UI and and working my way around there and that was fun those those were good yeah yeah yeah I was very proud of that I had a spark 10 on my desk for a while you know and so that was that was always pretty cool but yeah I was Linux and Unix and but I was a Windows programmer for many years too so you know who knows so what are you kind of most excited about with kubecon is this this you've been to other kubecon yeah I've been to I think this is my third or fourth I'm not okay I don't even know I lost count but I I'm I mean I was always excited about just talking to some of the contributors I mean I there's some people I mean every time I come here I there's some people like I know that I'm you know I'll see but then there's always some other people that you just you haven't and you kind of run into them and and meet them for the first time with her when you see there right on one of the talks and so that's always cool to see and there was some there's a few people that I was glad to have conversations with here yeah yeah nice um yeah that's cool and uh so how long are you staying for longer than the conference are you just flying home on Saturday or Friday I'm a flying home on uh Friday okay um are you did you get to see the tools no I didn't yeah I someone has told me that I need to go but I don't know if I'm gonna make it though because I'm flying out tomorrow morning and you're supposed to go in the morning because apparently it's Pat yeah but I saw some pictures so I'm like yeah it's supposed to be kind of a sight to behold but I'm kind of the same boat oh I'm not flying out till I'm flying out Saturday afternoon so I was thinking about trying to see if I could do it tomorrow at some point um but there's also like it's like a scheduled event like I can't just you know like roll up you know when I have a free minute I have to go and like it's a commitment um and you know when traveling especially for a conference making commitments too much of anything it's very difficult yeah um you know I end up like I'll I'll often have this long list of talks I want to go to and then not get to a single one you know yeah um so that's cool um and you said you're in Connecticut what do you think um you know you're going back there and then uh you'll go to the next cube con are you gonna go with anything in between yeah I probably the next cube con I I really want to submit a talk um I'll probably do um another so I so this set was an user panel uh I at some point I'll submit a talk and talk about like a deep dive about like our infrastructure at um when you first now at some point I'm sure people will be excited to see that so I at some point will um submit a talk for it but I I'm just trying to find the right the right time it's going to work with everything with my schedule and everything that's going on but yeah um so either in Chicago um or the next one and you wherever that is yeah yeah we can look for that Chicago is a good town so yeah I've never been oh yeah my my brother lives there so I've been a few times um but uh yeah it's uh it's a good place to go it's very it's funny how different it is than Boston because like um all the streets are really wide um and you know the uh you know most people like follow the driving rules and you know nobody crosses the street without a light um you know it's just like all the you know because I mean I assume you spend a lot of time in and around Boston oh yeah well see this is the thing in New York City it's the same way like yeah there's no rules when it comes to like crossing roads and stuff so I think some of that got into the Boston culture yeah well you know supposedly and I believe this to be true um but uh the term jaywalking was actually invented in Boston because there was an advertising campaign by the auto industry um when because traditionally right roads were actually for people not for cars and when the cars started to come out um they wanted to get people into sidewalks and stuff so uh they did an ad campaign of look at those jays walking in the street and jays being like fancy people and that's where jaywalking comes from supposedly so um well so why don't we say uh you know thanks so much for coming uh it was really lovely to talk to you I hope we can do this again sometime maybe in Chicago maybe somewhere else and um you know thank you yeah my pleasure it's fun and hopefully you enjoyed your little tour of Amsterdam um you know some parts that we hadn't seen before yeah the city parts we almost made the tulips yeah probably yeah I don't even know what direction there is so um we should have thought about that as our route which is like just drive out to the tulips and come back again um that would be cool