Today’s episode features an interview between Matt Trifiro and Mahdi Yahya, CEO & Founder of Ori Industries. In this interview, Mahdi articulates his ambitious vision for bringing cloud-native technology to the edge and building a new kind of cloud with the ability to power and scale the new internet.
Today’s episode features an interview between Matt Trifiro and Mahdi Yahya, CEO & Founder of Ori Industries.
Mahdi is an entrepreneur, investor, and former telecommunications executive who spent over 10 years building networks in data centers across the globe.
In this interview, Mahdi articulates his ambitious vision for bringing cloud-native technology to the edge and building a new kind of cloud with the ability to power and scale the new internet.
Key Quotes
Sponsors
Over the Edge is brought to you by the generous sponsorship of Catchpoint, NetFoundry, Ori Industries, Packet, Seagate, Vapor IO, and Zenlayer.
The featured sponsor of this episode of Over the Edge is Zenlayer. Improving user experience doesn't have to be complicated or expensive. Zenlayer helps you lower latency with on-demand edge services in over 150 PoPs around the world. Find out how you can improve your users' experience today at zenlayer.com/edge
Links
[00:00:00] Hi, this is Matt Trifiro. CMO of edge infrastructure company, vapor IO, and co-chair of the Linux foundation, state of the edge project. Today. I'm here with Madhi Yahya founder and CEO of Ori industries. We're going to talk about Mahdi's career in technology and his mission to build a new layer of flexible software driven, intelligent infrastructure.
[00:00:18] Hi, Mahdi, how are you doing
[00:00:19] today?
[00:00:20] Mahdi: [00:00:20] Hi, I'm doing okay. How are you doing?
[00:00:22] Matt: [00:00:22] That's terrific. little, Little fatigued from the COVID lockdown and the kids at school.
[00:00:28] Mahdi: [00:00:28] Yeah, I know. I understand. It's everywhere. I
[00:00:31] Matt: [00:00:31] you? Where are you dialing in from?
[00:00:34] Mahdi: [00:00:34] I'm based in London.
[00:00:35] Matt: [00:00:35] Based in London. Okay. And, has already, industry's always been headquartered out of London.
[00:00:39] Mahdi: [00:00:39] Yeah. Yeah. So when, we, started the company about two years ago. Yeah. It was started that from London, I've been based in London for about 10 years now. So a lot of the work
[00:00:49] Matt: [00:00:49] let's go back in time a little bit. So how did you even get into technology?
[00:00:53] Mahdi: [00:00:53] Okay. technology in general, I think I was the first kid in school that had a computer actually. So that was my [00:01:00] first, you know, journey into technology. And I was always been obsessed and interested in tech in general. I used to take things apart, take all the toys apart, see what's in them. and then I was the first kid that I actually had internet in school.
[00:01:11] So. I've always been involved in the whole internet and technology world from a very early age. actually made a killing from just renting my PC at home to the other kids in school to have
[00:01:22] Matt: [00:01:22] Well, it was an IBM PC. Your first computer
[00:01:24]Mahdi: [00:01:24] it was a Compaq.
[00:01:26] Matt: [00:01:26] compact.
[00:01:27] Mahdi: [00:01:27] Yeah. They're quite decent machines.
[00:01:29] Matt: [00:01:29] And what do you mean renting it? You were, you had a cloud back then.
[00:01:32]Mahdi: [00:01:32] no. I used to rent time for kids because some kids never seen internet before. And some of them has never sat on a computer before. So I used to have video games on my ms. Dos and, you know, I used to charge them per the hour to sit and play on my computer now. So
[00:01:46] Matt: [00:01:46] So always been an entrepreneur.
[00:01:48] Mahdi: [00:01:48] Yeah. It was fun.
[00:01:50] Matt: [00:01:50] And when did edge computing come on your radar? When did you first start thinking about edge?
[00:01:54] Mahdi: [00:01:54] Ooh, I mean, edge, I think it's something I stumbled into from [00:02:00] my work in the telecommunication industry at one point. but I started going into edge, probably around four, four or five years ago. I can talk about, how we actually got into it. I mean, there's a long background story to be honest in
[00:02:12] Matt: [00:02:12] Yeah, and I'm super interested. I know that, prior to founding Ori, you spent, quite a bit of times, quite a bit of time, operating a telecom and networking company in Beirut. And as I understand it, so maybe we could start there and you could lead into the evolution of Ori.
[00:02:27] Mahdi: [00:02:27] So I can talk about how I got into the telecoms world actually to begin with. So in, by roots, especially after the civil war, ended, and that's when I was grown up at that period, there wasn't. infrastructure yet in place. So people didn't really have landlines, then they didn't definitely have more buyers.
[00:02:46] So there used to be something called a cold shop and every, neighborhood used to have a cold shop where people who needed to make phone calls, they used to go to that coal shop and make a phone call. And naturally, because you're coming out of a civil [00:03:00] war, there was a lot of people who. Traveled. So there was a lot of people in the U S a lot of lemonades in the U S lots of people in Canada, in Europe, in Africa.
[00:03:07] So there were so many relatives that needed to talk to each other. So these phones were like, these KOLs shops used to be quite popular and it was a good business. my father owned one of them. So, I started working there as a teenager.
[00:03:21] Matt: [00:03:21] like the precursor to the internet
[00:03:23]Mahdi: [00:03:23] yeah, exactly. But it was a place, you know, they had a counter, you walk in and.
[00:03:29] Somebody gives you a phone number on a piece of paper. I say, can you doll this my son in Canada, I'd like to talk to them. and then you should basically dial the number for them and you have the time. So you put the timer on when people pick up, and there used to be this little phone booth on the side.
[00:03:43] So you can transfer the call to one of the phone booth and say, hi, you can go and talk. And number one, number two, it was quite fun. but obviously for me, I was really interested in saying, okay, how does this work? So. I know, like I can dial the number on this device in front of me, but how does this whole thing [00:04:00] work?
[00:04:00] Right. How to, how do we connect all these people? And that was the beginning of my journey in telecoms and networking and all of this world,
[00:04:07] Matt: [00:04:07] What year was this approximately?
[00:04:09]Mahdi: [00:04:09] this was a 90, 94 95, I'd
[00:04:14] Matt: [00:04:14] okay. So the internet was around, but the browser was still a science project
[00:04:20] Mahdi: [00:04:20] I know it took some time. to come to Lebanon, all of these things because of the infrastructure was not there.
[00:04:25] So even though it was a few things wet around in the outside world, like in other places of the world, but it took some time to come to Lebanon because the infrastructure was still being built. I mean, the civil war ended in 1990 in the nineties. So 1990, but this is what I started understanding that, okay.
[00:04:41] You deal with these outside companies that supply you with lines. and one of the companies that, you know, my dad's coal shop used to have a contract with that, supply them with satellite connectivity so that you can make international calls was actually based in London. So on, I came with him on a business trip to meet these companies.
[00:04:59] And this was the [00:05:00] first time I'd come to London. So I fell in love with the city since then. That's why I moved here. but also it was the first time I go into a company it's an actual company with people and desks and phone calls and numbers and papers and people in suits. and I was, I also was obsessed with that and I was like, this is so exciting.
[00:05:17] What is this thing? Like, what is, you know, it's a company, And, since then, I thought, you know, when I, when the time comes, I'm gonna start one of these companies, these carriers, and that's what I did. So, Sam, my, the first company was a wholesale carrier that does exactly that. And that was the starting point for me.
[00:05:36]I was about 21 years old, and I decided to start a company like the one I saw when I was a teenager.
[00:05:43] Matt: [00:05:43] But that is really, that's a great story. I mean, yeah. to have such unbounded ambitions and the sort of presence of mind to invest the time and energy to make something real at that age 21. I know what most people are doing at 21. Most of my [00:06:00] peers. That's really terrific. And so what was this?
[00:06:03]w what's this company, I mean, just like the com was it called? What is it still around?
[00:06:07] Mahdi: [00:06:07] Yeah. I mean, it's still around to this day. I'm not part of that company anymore. I'm not involved in it. but it was, what you call a wholesale carrier. So you build pops around the world, and you connect them with, you know, fiber and multiple ways of connecting them. And you basically move traffic between different networks.
[00:06:26] So you sell wholesale. capacities to networks. we primarily were focused on voice traffic and SMS traffic. So we're basically shifting a lot of voice investments, traffic through our routes and our networks, and we became really good at it. And you supply routes to mobile operators, to fixed line operators, to retailers like coal shops as well, who don't have access to.
[00:06:46] So you basically supply them with these lines and they're able to make phone international phone calls. And then through the network, you route these phone calls, whatever they're going to the world. So there was a lot of intelligence that had to be built within the network as well, that allows to know, okay, this person [00:07:00] is calling the U S this person is calling this network in London, for example, and so forth.
[00:07:05] And you charge for them. but at one point we became, one of the largest in the middle East actually doing that. So we were moving. About a billion voice minutes a year at one point, and over about 3 billion SMS is through our roots. And that business is what basically taught me what the data center is and how data centers work and how they send us get connected and peering and interconnection and all of this world that is not a big part of answer as well.
[00:07:31]but that was my start in this space. And I did that for about, so seven years of my life and has not company.
[00:07:38] Matt: [00:07:38] Yeah, that's really quite a story. So, so you built this telecom network and I want to come back, we'll come back to talk about
[00:07:45]Mahdi: [00:07:45] yeah.
[00:07:46] Matt: [00:07:46] you're right. It is very important at the edge, but I want to, I want our listeners to understand what Ori is.
[00:07:51] So in 2017, you've founded ori what, what led you to the founding of Ori?
[00:07:55] Mahdi: [00:07:55] So in the telco telecom space, I'd say about [00:08:00] probably seven years, seven, six years ago. There's been a big shift in the telco space. And, you know, especially in if field in the voice business, you will be making a lot of money at one point. I mean, the company we're making an access to 200 million revenues per year and you think life is good.
[00:08:16] People are not going to stop communicating people, not going to stop making phone calls. Right. Who would think that like this people are
[00:08:22] Matt: [00:08:22] I can't remember the last time I made a traditional phone call. Yeah.
[00:08:25] Mahdi: [00:08:25] Right. I know. I know. but then suddenly, and I mean, you know, if these stories, right, you've seen that everyone, and everyone talks about it, you know, all these revenue started dropping not yearly month a month and he'll like, f***, what's going on?
[00:08:39] Right. What happened? I don't, can I swear on this? I'm sorry.
[00:08:42] Matt: [00:08:42] yeah, we'll figure it
[00:08:42] Mahdi: [00:08:42] Yeah. And, you know, so I naturally me being curious about what's going on and how these things are happening. how's the world changing? I looked around and there was two giants that were growing around us, right? The public clouds from one side.
[00:08:58] They have all of these [00:09:00] applications. I mean, not just the applications do today, but all of these OTT applications as they call them over the top applications, you know, like the WhatsApps of the world and Skype and Viber and we-chat and hundreds of them. There's so many of them now we'll do like a voiceover IP connectivity that does not really need all of this infrastructure of networks.
[00:09:17]and the most ironic thing is that. We as the teleco industry invested overlying $4 trillion in 4g infrastructure, thinking that this is what's going to be the next thing. And this is what's going about. This infrastructure actually powered all of that. So the market's shifted the market power, moved to these two different sides, but at the middle, the cost of maintaining all of that stayed the same.
[00:09:42] And that affected all of us. So for me, I actually stood there and was looking and I looked at the site. And so like, there's hundreds of these applications that are there and if I'm going to be doing something else, I don't think I will go into that space and compete with the WhatsApps of the world or anyone like that.
[00:09:58] But then I looked on the right side and there's [00:10:00] like three or four companies only who are doing this and own this big market that is growing around us. And obviously edge was. Starting to surface people talking about edge, mean people talking about edge for a very long time, but people started talking more about edge with now, with 5g and everything else.
[00:10:15]and that's where I started thinking about, okay, with cloud is not going to end that there's going to be another chapter for that. and it's only going to grow because of the way things are going because of all of these applications. And it's going to become massive and big, and it's not going to be on its own.
[00:10:30] It's going to need this infrastructure that we're building now, we're building next. so that was the beginning of my thinking about how can I utilize, why I have not, what I know now in this industry, in the telecommunication industry and use all of these cloud stuff, the cloud native stuff that are growing on the side and create something new.
[00:10:47]and that was the beginning of my thinking for already. It took some time to actually realize, okay, there's and how can we do it? How can we apply, cloud native technology to the edge virtualization of networks and so forth, and then put it all together [00:11:00] into a company we call today Ori
[00:11:02] Matt: [00:11:02] Industries.
[00:11:03] So let's, let's walk through that. I mean, what, w what is your product? what is it, what are the parts of it and who uses it and how do they use it?
[00:11:12] Mahdi: [00:11:12] So on the surface,
[00:11:18] there's a new internet today, right? And you talk about it specifically. I've seen you talk about it before. I've seen you write about it, the new internet, And then there's three, there's three main categories in this internet. Now we call it there's an autonomous internet, right. Machines that need to make decisions on their own machines that need to communicate quickly with each other.
[00:11:36] Right. that is, An immersive internet, where it's virtual worlds that we interact with on daily basis. and there's a smart internet like hospitals, cities, factories that are extremely intelligent on autonomous today. and this internet requires, a new cloud. To deal with and a new cloud to power and to be able to deliver and scale it.
[00:11:57]so as Ori we position ourselves and we are actually, [00:12:00] we're building this new cloud that will allow us to power and scale this new internet. Now as a product, we are, we're a cloud. We have cloud products on top. We allow application developers to deploy. across thousands of edge computing nonstop.
[00:12:15] We host in mobile operators in ISP, in telco networks, in edge data centers in Metro data centers at the moment. And we are building a fabric, which we call Ori global edge or OGE that allows us to provision all of this infrastructure, allows us to add a lot of intelligence in it and allow us to build products on top, whether it's a container orchestration products, VM products, or functions, or even later on.
[00:12:39] Multiple different products that we're thinking of allows us to basically start moving all of these products across one unified fabric over all of this infrastructure that we're building.
[00:12:50] Matt: [00:12:50] Got it. So if I can unpack some of that, tell me if I've got this right. So when we think about a cloud, we think about, being able to go to a dashboard or to [00:13:00] programmatic programmatically request resources that are global. Right. And in the case of like Azure or Amazon, I say, I want to, I want to, easy two instance in us West and NEC to go for instance, in U S East and you know, a couple seconds go by and I've got two virtual machines or whatever I ordered.
[00:13:17]and it sounds like what you're doing is something very similar. but you're doing it at thousands of locations, is that in those locations are much farther out in the network. They're not in these large centralized data centers. They're out in, as you said, Metro data centers in, you know, adjacent to the access network, like really in that, service provider zone that we talked about it at the Linux foundation.
[00:13:38] Okay. Okay.
[00:13:40]Mahdi: [00:13:40] and we actually, we to Tibet to allow us to actually tackle all of this because as you know, this is a huge
[00:13:46] Matt: [00:13:46] That's pretty ambitious. Yeah. I want to ask a lot of questions about
[00:13:49] Mahdi: [00:13:49] filled with snowflakes everywhere. And it requires a lot of work, a lot of thinking, and it requires a lot of automation and that's a big thing that we constantly try and deal with, ori is that we need to automate as much as [00:14:00] possible so that we're able to access all of them.
[00:14:02] We add able to actually provision it in a right way. but we divide it into, there's something we call a Metro edge cloud and I think the name defines it. And it's basically everything that is in Metro data centers or in multiple certain areas. and that's a layer we're building at the moment.
[00:14:17] So we're actually placing servers in Metro data centers. We're primarily doing that in Europe so far because we're based here and, we've had work with EdgeConneX and we're having worked with, with new data centers now where we're placing, in strategic places. The second layer underneath that, we call it a central edge cloud and that's primarily in, telco central offices.
[00:14:38] And this is where we start moving from the edges of the public internet and going into the telcos, the ISP and everyone who's.
[00:14:46] Matt: [00:14:46] And for those people that don't know, a central office is essentially a historical aggregation point, in a Metro area. And, more often than not, they're a little closer to the access edge and a re than a regional, but it depends on the geography. I mean, sometimes they're like literally [00:15:00] next to each other, but it is part of the historical telco infrastructure.
[00:15:03] Mahdi: [00:15:03] Correct. And the interesting, and the reason we're able to actually utilize this today is because a lot of them are being transformed into actual micro data centers. So power, connectivity, cooling, whatever you would find in data centers, telcos are actually doing that today. And primarily for. 5g and the adoption of 5g and, and virtualizing of the networks.
[00:15:24]so there is a big chance for us to actually put generic compute servers there and run different functions in there rather than just network functions as they're doing.
[00:15:32] Matt: [00:15:32] Yeah. And the other important thing, and I didn't mean to interrupt you, naughty, but the other important thing is that, those are, backhaul aggregation points for the telecom
[00:15:41] Mahdi: [00:15:41] Yep.
[00:15:42] Matt: [00:15:42] means that you, which is really, you know, you think about the edge, you know, we like talk about the edge is as much a network problem as it is a, where do I put my server or my storage problem.
[00:15:51] Mahdi: [00:15:51] I mean, it's definitely a network problem that is being solved today. Right? It's cause placing servers in place, in multiple locations you can do that. They, and you can [00:16:00] provision or you can rent stuff on the bus. How do you connect them all together? how do you network, how do you dynamically route things between them is a big challenge and it's the exciting challenge about agile?
[00:16:09] I think.
[00:16:10] Matt: [00:16:10] So you have the Metro cloud and the central cloud. And there's a third one you said.
[00:16:14] Mahdi: [00:16:14] And the third one is what we call a local cloud. And that's what sits on the access layer. Of the tacos and that would be the one that will happen in the future. It's not fully happening now. and that is dependent on where the telcos are when in the strategy virtualization adopting virtual ran radio access networks and adopting Oh random and all of these different technologies that are coming up now.
[00:16:37]but that's the furthest layer you can get to in the network before you get to the user devices, and we call that a local edge cloud. and we're starting to. Build these different layers in phases. So we've started with a Metro edge since last year. we have six major locations in Europe today that we're running, customer workloads on, we've been running a lot of pilots [00:17:00] recently and for quite a long time though, actually, but in telco years, it's a very short, like recent time, in central locations.
[00:17:07] And the third layer is roadmap for us to start running that once the infrastructure is ready, we call this the edge gradients. and, there's multiple qualities of service, obviously on each of these layers and in each on of these gradient, there's multiple types of hardware. You can put that because the footprint gets smaller as you go down the further, down the edge of the teleco
[00:17:28] Matt: [00:17:28] does. Yeah.
[00:17:29]Mahdi: [00:17:29] and it serves different applications.
[00:17:32] So Ori global edge OGE is actually the fabric that allows us to provision all of this gradient allows us to start moving workloads across it in a dynamic way and allows us to move things up. The gradient scale for example, applications are quite compute heavy and for other applications that are.
[00:17:49] Bit less, more nimble that require a more agile can sit on the edge or require different quality of service that sits on the edge. Not necessarily, I'm not just talking about latency. I mean, latency is a big factor of that [00:18:00] and we measure latency as we go up to the gradient or down the gradient, but there's different applications that we're working with today that are not really latency sensitive, but they're more, Geographic placement of data sensitive, for example.
[00:18:12] So where do they want to place the data? Where does the data need to be processed within a certain proximity of where it's being generated? And that's quite, I think is an interesting use case use of here recently. I've we've come across in age. so our platform allows. This dynamic orchestration across all of this gradient.
[00:18:29] And we can start basically offering applications, multiple locations based on what they need from a quality of service, but also financially as well, because the more you go down the gradient, the less than at what cost and set away, and you're not paying for back on. I mean, you were talking about. Well, central office has a backhoe.
[00:18:45] We can reduce costs on the back haul. If we move things down further down to the edge, obviously you'd have more cost and compute and different places, but the back-haul becomes less expensive in that space. And that's what we're trying to tackle at the moment.
[00:18:59] Matt: [00:18:59] Yeah. And [00:19:00] what few people realize is that if you look at the network infrastructure that connects those three layers of the gradient, each of those backhaul links is oversubscribed and yeah. And the tremendous amount of congestion. And so just pushing something out that reduces the amount of.
[00:19:20] The amount of network capacity required in the Metro region to process the data is huge. Yeah. So, so let's talk about a few things. Okay. So, so this gradient, okay. This gradient that spans the region, the market. The metropolitan market and you're divided into these, I think really clear, distinctions.
[00:19:38] And obviously they overlap. there's, you know, there's a lot of fuzziness there, but each of them will have a distinct, service level. And I appreciate that you pointed out that it's not just latency because I think a lot of us in the edge industry and I was guilty of this really harped on latency, but at the, you know, the span of.
[00:19:54] The sub 50 miles sub a hundred miles, you know, latency is often negligible. [00:20:00] you know, the time it spends like writing to the district or processing it. So there are some applications require really low latency, me driving the remote radio, head in the 5g network from a virtual function, for example, it's, you know, it's very related to sensitive, but a lot of things aren't.
[00:20:13]as latency sensitive. so I appreciate you bringing up sort of some of these other concerns. So let's talk about the data example. so what's, can you describe like a real world use case that's generating data where it's important to keep that data, you know, off of a large part of the backhaul network and process it in the low?
[00:20:30] Is it the local zone, the local
[00:20:32]Mahdi: [00:20:32] local, even central office, depending on.
[00:20:34] Matt: [00:20:34] central cloud. Yeah. So how, so give me an example of one where that's an important distinction to make.
[00:20:38]Mahdi: [00:20:38] so, I mean, in Europe we have something called GDPR, for example, at the moment. So there are certain organizations, like healthcare organizations or hospital organizations within regulations. They really can't move the. Patient data or the data outside the premises or within a certain kilometer radius of where that data is generated or entered and stored [00:21:00] as well from.
[00:21:00] We mentioned. So, a lot of them have to invest in. On premise, hardware and infrastructure so that they can do that locally because they can't be saved in a third party place or in these clouds and public clouds as most people do to this day. and that's actually a real life use case we worked with, as part of a, there's something called innovative UK here in the government where you basically work with.
[00:21:25] You know, government organizations and government has funding for those. and we did a bit for that. You do R and D projects within that space. and that was our, our role was working with the local telco who has an edge node that is within that geographical location where this data needs to be saved and stored and access to whenever it needs to be.
[00:21:44]it reduces costs on. These health care organizations from running on-premise infrastructure from building an actual server room, from hiring a, an it infrastructure team that needs to run that. I mean, you know, and putting all the [00:22:00] software that needs to do that and everything else. and then basically they can just focus on what they're doing and.
[00:22:05]the same service they would get from being one on premise. They could have that as a service from an edge cloud that is in the region within the local telephone there. and we were the platform that was basically, facilitating that. And that's why I say it was fascinating thing that I didn't think of until we started working on it.
[00:22:23]Matt: [00:22:23] it's a big trend in the U S we've started calling it near prem.
[00:22:26] Mahdi: [00:22:26] Yeah. Yeah.
[00:22:27] Matt: [00:22:27] on-prem and there's near print, right? Cause you're, if you're, you know, if you're 20 or 30 kilometers from the hospital, you probably can conform to the GDPR guidelines or the policies of the facility, but you're right.
[00:22:40] Take all that cap ex out. Take a lot of the op ex out and have the, sort of the capabilities of cloud, you know, some on-demand scalability, you know, things like that. So that's, yeah, that's a really, I was surprised by that use case too. Although in retrospect, it's sort of obvious, right? It's like such a huge value proposition.
[00:22:58] Our CEO, Cole Crawford, you [00:23:00] know, it's a little flippant, but it's kind of a fun statement and I think there's some truth to it. He says, look, the killer app for edge is a cost and the easy button. And essentially to some extent that is what the Ori global edge cloud is, right. it's, you know, it's a lower cost way of getting, you know, solving for these edge compute problems, whether it's latency or local localization of data or whatnot, you know, w and to do it easily.
[00:23:24] So, okay. So you're building all these cloud-like abstraction layers across these three, this gradient, and all of that. And at some point, you know, Your customers might not see a lot of this, but when you peel the Popeil, the covers off there's hardware underneath that, what are you deploying that hardware?
[00:23:41] Are you partnering with other companies that, you know, bare metal providers or leasing companies or something? how has, how are you actually getting your cloud up and running globally?
[00:23:51] Mahdi: [00:23:51] three locations, we are, we're deploying the hardware and we run our own MPLS network across all of these locations at the moment. we have a fantastic team that [00:24:00] does that. We've been doing that for a very long time. I think you're interviewing one of the guys who runs this team next week. I think we
[00:24:07] Matt: [00:24:07] Super. I'll have a lot of, I'll be knowledgeable.
[00:24:11]Mahdi: [00:24:11] and he did that for one of the hyperscalers. So he brought in a lot of that knowledge and he's helping us scale a lot. And he built the infrastructure team, that is basically looking after all of this, on the central and local layers. It depends. On the discussions and the partnerships and the agreements we have with each local telcos.
[00:24:30] So some of them, as you know, have already in edge strategy, some of them are deploying edge. Some of them are working with certain partners on edge and basically building edge clouds. some of them are early in their journey, so it depends on where they are in the journey. How much investment are they willing to make into this?
[00:24:46] We can bring our own. hardware into there. And for us, it's the same as a Metro location. And we just connected to our network and we partnered with them on different other things. if they have their own hardware, then we also use that. and basically our OJ, [00:25:00] your fabric can talk to that and provision that hardware and add it to the rest of our cloud as well.
[00:25:05] Matt: [00:25:05] Okay. Well, that's a perfect segue into my next question, which is, so let's, let's fast forward and imagine you've got, I don't know, a hundred global telcos as partners. Okay. Each of them was, as you point out a different edge strategy potentially. Even if they're like similar, they're still different.
[00:25:24] How. How do you make that easy for your customers to consume? Because somebody might have a GPU heavy strategy and somebody else might have a, you know, I don't know. So I guess to tell me about, you know, the word is federated. And I talked to you before the show about defining that, but yeah.
[00:25:42] Why don't you walk us through how you view that world and what kinds of tools you see bringing to your customers to help make that easier to navigate?
[00:25:51]Mahdi: [00:25:51] so for application developers, as customers, they don't have to worry about any of that. So they don't deal with this. They really don't know what's taking underneath. [00:26:00] And that's part of our proposition. You know, you deploy in a cloud and we take care of everything else. working with telcos and we do actually have a product for telcos that we've been working with them on.
[00:26:12]and it's a product called helix. we haven't announced it yet actually. We'll be announcing this soon, with our developer facing product is actually called edge crane. And we'll announce that soon well, but he makes them a product is, is prominently a slice of OJ that basically get we give to the telcos and it allows them to build local edge clouds and these local edge clouds gives them the tools to sell to their enterprise customers because working a lot with well, because we realize that they have a lot of enterprise customers that.
[00:26:40] Matt: [00:26:40] they have the sales team
[00:26:42] Mahdi: [00:26:42] Exactly. They have the status things that we don't have yet, but they also have large the price customers that would benefit a lot from edge. and for us to build the business case for them, we have to actually show them how they can sell that to their enterprise customers and how they can increase revenues from the existing, under price customers, through selling as [00:27:00] solutions.
[00:27:00] So we built them this product, which allows them. Effectively to build a central or local edge cloud or both. And part of the work we're doing with them is that we're helping them build these centralized clouds. We're giving them the tools to actually sell it services like that. Any cloud, like a local cloud in the, geographies, but also part of the partnership is that it gives us access to OJ or a global edge to actually deploy to these local clouds.
[00:27:24] So by doing this, we're kind of. Unifying a bit on we're standardizing a bit how these local edge clouds are looking, how do these global edge clouds interact? So it makes it easier for us to eventually link them all together and work on top of them or work with them to basically deploy applications from global developers to that.
[00:27:44]We talk about Federation, for us Federation is number of helixes within different telcos. So a helix in, at and T could talk to a helix in BT, in the UK and a helix and Telefonica can talk to a helix and Vodafone, as, as a product, as a [00:28:00] platform. and we create converged API between all of these multiple, local clouds that the telcos are building, because we said, look, Even if your customer wants to reach global reach, then your customer can have the capabilities to deploy applications in BTS network.
[00:28:16] If you're a customer of Vodafone through these platforms, token together through conversion APIs at the moment. and this is exactly what I used to do in the wholesale world and the telco side, because we use the federal grants, right. I, a customer. Calling a customer, a MBT calling a customer in at and T in the us, we had to federate between them so that these customers be able to make phone calls and talk to each other.
[00:28:40] So this is the idea. This is where the idea came from in the first place. And I said, why can't we do that with cloud workloads, with edge workloads? Maybe we can do that. So a platform and BTS network can actually make a request. In, through the platform in, at, and T's network and deploy your workload for the customer and, for BT and [00:29:00] within that Federation, there are, you know, billing, there is settlements, there's all of these different things that we used to do in the wholesale world anyway, and an automated way.
[00:29:08]and that's what Federation is for us. So all of these different platforms and different tacos can actually talk to each other and create a network between each other and serve customers globally. Even they are local customers for local telcos,
[00:29:22] Matt: [00:29:22] Really clear that you've built something analogous the past, because you mentioned billing and settlement.
[00:29:27] Mahdi: [00:29:27] of course. And that's why I
[00:29:29]Matt: [00:29:29] there isn't a developer on the planet that thinks about that or wants to deal with that,
[00:29:33] Mahdi: [00:29:33] they don't have
[00:29:33] Matt: [00:29:33] one of the most. Yes, exactly. Right?
[00:29:36] Mahdi: [00:29:36] Well, that's why I say I stumbled upon edge because of where I was and because of the history of war I'm in the telco space, you know,
[00:29:43] Matt: [00:29:43] Yeah. So, so, okay. Let's talk about the global developer. Cause I think that's one of the most exciting things, which is, you know, particularly for, smaller countries that have maybe one or two, telco networks that aren't large enough on their own to attract some of the, [00:30:00] you know, bespoke programming for whatever, whatever unique. ed strategy they have. And so the ability to develop for the OGE platform and, simply deploy my application and have, you know, your system do the hard work of figuring out, well, how do I run it here? So how does that work? I mean, how do I tell Ori what resources I need and what if those resources aren't available at the level that I'm requesting them?
[00:30:26] Like, how does that all that magic work.
[00:30:28]Mahdi: [00:30:28] so, I mean, it's like any other clouds, so you can define regions. You can say, I want my application to run in these geographies, in these regions, or I wanted to run globally, whatever OGE has presence, please run it there whenever it's needed. at the moment we, we offer containers so you can run a container across our platform and we basically deploy it.
[00:30:47] Wherever it's needed. the unique thing that we're building in the platform, that's something we've, we spent a lot of resources on and we're still are, we're building a lot of intelligence in terms of, eventually we should be able to [00:31:00] tell where to place the container at the right time, even before sometimes the request is coming.
[00:31:04] So when requests and demands start coming from users, in certain geographies, then we spin up a container. In the right edge location that can confer, comply with, the different manifests that I develop assess. So I said, I want this type of in latency. I want this quality of service. I
[00:31:21] Matt: [00:31:21] I see. So there's a declarative manifest that says, you know, like I, this is a hospital workload. I need to be within a hundred kilometers of this lot of latitude, longitude or something.
[00:31:32] Mahdi: [00:31:32] Exactly. And we basically spend that container up wherever it's needed, based on this manifest that the customer, the, the fines. now there's some like enterprise customers that we're starting to work with today. I mean, they don't have manifests or anything like that. So they have certain requirements that we put in, but we're working more and more towards make it even more intelligent so that eventually you don't really need to give us a manifest.
[00:31:55] You know, we'll place it in the right place based on the type of application that you're running, [00:32:00] whether it's an IOT application or a gaming application, or as you said, as a health care application. And we know all of that and we'll be able to place it in these right locations whenever it's needed.
[00:32:10] Whenever the demand is coming from the users.
[00:32:14] Matt: [00:32:14] Yeah, that's really cool. And so, you mentioned that, you know, you're on the timeline where you've started to deploy the Metro cloud, you know, what do you see the. from a time perspective, where you start, you know, fleshing out the different cloud layers.
[00:32:30]and when do you think that the gradient will be complete in, I dunno, the top world.
[00:32:34]Mahdi: [00:32:34] so we were going to start having, central cloud grading ready for production, mid next year. So that's when that's the target. So in June we are having, central locations already for production, not as many as we want, but we will have a few where we can start moving some of our existing developer workloads, Metro to these locations, on the local level, I don't think that's going to happen before 20, 23, to be honest. [00:33:00]
[00:33:00]and that's basically because of, the timeline for adoption of 5g, because that would lead to the adoption of, you know, distributed core, that will lead to the adoption of, Oren or viron, in telecommunication networks, which will allow us then to build local clouds there. we've run a few tests and pilots at the access layer with a few telcos, but it's still very early on because.
[00:33:23] It's not just about placing the computer underneath the base station of some people imagine, as you said, there was a lot of networking involved to be done. And there's a lot of integration with the network, with the user plane traffic, because being in a network allows us to basically get a lot of data about how the users are reacting, what are the users and everything else.
[00:33:42] And we need to make use of that so that we can build more intelligence into our platform as well. and the local edge layer is all about that. so there's much more to be done yet on the network layer before we can even just place service that on ground production workloads in that space.
[00:34:00] [00:34:00] Matt: [00:34:00] Okay. a couple of other questions. I have a lot of questions. So, so, so you call yourself a cloud provider. there are other cloud providers, you know, Amazon, IBM, Oracle, Google, Microsoft, how do you see Ori industries relating to those other clouds?
[00:34:18]Mahdi: [00:34:18] I mean, we're just like another cloud. Like them, it just happens that they're much bigger at the moment.
[00:34:23] Matt: [00:34:23] so.
[00:34:27] I, you know, I just love the fact that, you know, I started to telecom network. I'm going to start a cloud.
[00:34:32] Mahdi: [00:34:32] Why not?
[00:34:33] Matt: [00:34:33] I it's it's it's it's it's thrilling to hear you have so much confidence. It's really a neat,
[00:34:39] Mahdi: [00:34:39] But there is an opportunity, you know, if you know, telecom world really well. I mean, there is an opportunity right now and my focus is to seize the opportunity and I'm working with all of the telcos because I know them, I've worked with them for a very long time. I've worked with them from the outside, you know, in wholesale, in Federation and all of these things that we can apply to edge.
[00:34:58] So I'm trying to apply all [00:35:00] of the knowledge.
[00:35:00] Matt: [00:35:00] the customer a lot better than, yeah.
[00:35:03] Mahdi: [00:35:03] So, and it's working. I know it's not, it's been working so far and we've gaining a lot of traction. We're working with a lot of telcos. you know, we're part of this, group under the GSMs. Now that has about 44 telcos and we're defining standards there as well as toward
[00:35:17] Matt: [00:35:17] What is the group
[00:35:18]Mahdi: [00:35:18] it's the OPG, it's the operator.
[00:35:20]it's the, Operator platform group, that was set up on the . there's been announcements about it, everywhere. And we're releasing the first set of standards, pretty soon. and there is a willingness to work and to build a Federation and to do all of these things that we are actually pursuing and suggesting.
[00:35:36] So I'm quite optimistic that we can do this actually.
[00:35:40] Matt: [00:35:40] Yeah. and there certainly is an opportunity. I mean, the it's not gonna, the window is not gonna be open forever, but there is, there's definitely an opportunity to get ahead of people in the edge world. so let's talk more about the telcos. So, so for years, the, every time you heard edge computing in telco, you would almost hear.
[00:35:59]ETSI MEC, [00:36:00] which used to be mobile. It's computing now multi-asset is edge computing in the next sentence. how do you view, ETSI MEC and how do you relate to it? Okay.
[00:36:07]Mahdi: [00:36:07] I mean, it's, we comply with some of the ETSI MEC standards, on the teleco local level. So when helix gets deployed in a telco, you know, our telco product, there are elements of compliance with that because a lot of the telcos adopted these standards and they're adopting these. Dundas with, you know, how they did with VNS and VNF orchestration.
[00:36:28]so we do comply with that. we. we did not build our platform on these standards. There's elements of that we comply to that allows us to basically talk to whatever the telcos have. if it's compliant with ETSI MEC, I think it's a good initiative. I think it was a bit premature to be honest, to go in and just put all of these standards now for MEC.
[00:36:47]but I think it did actually opened doors and a lot of conversations for telcos into that space. and that's what I see the value from it primarily.
[00:36:54] Matt: [00:36:54] Well, and it's also evolved too. you know, originally it was really ambitious. Like it was going to be a [00:37:00] cloud, right. I mean, it came out of the Nokia liquid apps and all of that. And I think people realize that's just not. The telcos aren't going to be successful at it. So it's become more of an API, which I think is, is a really sensible way to go because this opaque telecom equipment, you know, to translate it into something that developers understand and cloud makers understand really makes a lot of
[00:37:18] Mahdi: [00:37:18] And it made our conversations with telcos easier because you know, it kind of, it was the
[00:37:23] Matt: [00:37:23] They're
[00:37:24] Mahdi: [00:37:24] production for them. Yeah. So I think that's the biggest value that they've done in this.
[00:37:30] Matt: [00:37:30] Yeah. So, so as telcos move to 5g, one of the biggest trends as you know, is virtualizing your network, you know, running these network, network functions on white box computers that could be cloud computers. Do you see a world where a telco could actually deploy its network functions on. On, you know, Ori global edge
[00:37:53]Mahdi: [00:37:53] I wouldn't dismiss it, but I don't think it's something we're pursuing at the moment.
[00:38:00] [00:37:59] Matt: [00:37:59] that's for year four. Yeah. That's I mean, it's really difficult, you know, logistically, politically and all of that, but, you know, it's all cloud workloads.
[00:38:12] Mahdi: [00:38:12] It's the same, it's an application, but there are people who do that today, and we're not, we don't want to try and compete with them. I think the value today is getting these developers because building an edge native developer community is a big thing, and we're all working towards that. And the answer, right?
[00:38:29] Because that's where we create the demand for the market. And the moment we create the demand for the market, then we can scale the market from an infrastructure side, from a platform side, for more than ever, we're doing what you're doing, what everyone else is doing. So the markets for VNS and orchestra, it's the, I mean, telcos are doing it.
[00:38:46] There's a few vendors who do that, and they're happy with that. That doesn't mean they might be happy for a forever, but they're happy with that today. And I don't think that's where the value of entry is at this stage. So our focus isn't to create that and not go after VNS for [00:39:00] now.
[00:39:01] Matt: [00:39:01] So, so if I was a developer or maybe, you know, a young woman or man in, a university and I wanted to become an educated developer, Because I watched my older peers become mobile developers and get wildly rich, you know, nine years ago. where would you suggest they go? What would you suggest they do?
[00:39:23]Mahdi: [00:39:23] the, to be fair, the resources are not that clear out there yet. And I think we have a responsibility to actually make that clear on and start growing that. But, I mean, you know, there's initiatives like LFedge at the moment and the Linux foundation that is actually working towards that, you know, the, state of the edge that you've you shared you've chaired and you initiated as well, which I think is quite a useful thing that, part of the LFedge now as well, but it was a good way to start that, developer community, as we start growing and not just us, a lot of different companies, there's going to be a lot of, material out there about how you build your application across us.
[00:39:59] And I can start [00:40:00] seeing that. No in the, in some of the applicants and some of the companies, documentations, and there's some blogs about that now. So this is what eventually people will start basically learning about agile. That's how the big clubs did it. We still need to demonstrate the value of the gradient because Metro is great.
[00:40:19] But once we start going out of Metro, that's where actually edge native will start because Metro is still it's on the border of cloud native slash originate. Right. and in order to do that, the technology needs to be a bit more ready. to be able to actually demonstrate the capabilities of, demonstrate the value of it.
[00:40:36] And then we can start communicating that and educating all of these new developers of what's
[00:40:42] Matt: [00:40:42] What, how do you see Ari's relationship to the wire line last mile? Well, so, so you're focused on telecom, which is a wireless last mile. what is your view, to the wired last mile?
[00:40:56] Mahdi: [00:40:56] Even the land line, we call it landline here. we work [00:41:00] with the mobile and line, like, so currently in.
[00:41:04] Matt: [00:41:04] Most of the companies are doing both now, anyway, at
[00:41:06] Mahdi: [00:41:06] Yeah. Yeah. I mean, it's, it doesn't make a big difference to us. And, some applications would run much better on Atlanta, landline, in broadband, for example, rather than, on mobile as well.
[00:41:15] So, for us, telcos are all of these networks. They're mobile they're landline. you know, we're working with a provider here in the UK, one of the biggest landline provider in the UK, and it's all on the, broadband, and London network. What they're trying to do with edge.
[00:41:29] Matt: [00:41:29] Yeah. You know, it's interesting. And I want to see if you have a theory on this is, you know, the, I think we, can you speak to the ecosystem about edge computing nine times out of 10, if you're not talking to on premises people, if you're talking to people that are looking at it from a cloud you know, full gradient perspective, Most of them are thinking 5g.
[00:41:51]and yet you and I both know that you can do this over LTE. You can do this over fiber. You can do it over DSL. You can do all these things. you know, a lot of the applications [00:42:00] that are counting on some of the more full featured aspects of 5g. So, you know, the massive number of devices, you know, network slices, things like that.
[00:42:08] You actually can do. Most, if not all of that short of mobility, in, you know, over fiber or other kinds of wireline, why do you think, w why do you think 5g has, seems to have been the catalyst for edge computing when it's been possible?
[00:42:26]Mahdi: [00:42:26] I think so there's a marketing element involved from the telcos. I'm from the vendors who are selling 5g. Right. Because there was a time when. Telcos hear about edge. and they say, okay, what's the value of fads for us. We're going to make more money or new revenues from new applications. Great. We want that, it's going to reduce our back haul costs.
[00:42:48] Fantastic. We want them, and it's going to make the experience for our users much better. Also. We want that as well. So there was a Mount marketing element, I think primarily from the vendors where they said, Oh, you want [00:43:00] edge? Adopt or adopt 5g because 5g will make edge possible, but also there's, there's something true to it.
[00:43:07] And it's, it has nothing to do with 5g being faster. So, I mean, internet is fast today. I can load a video easily and I can watch it with not interruption. Well, I think it's a technical element where, you know, we were talking about virtualization and network virtualization, tanning, a lot of these central offices into micro data centers of the network.
[00:43:25] And primarily this investment in virtualization is because five because of 5g, because 5g drives fire virtualization because you know, you start running network functions through software over. Genetic hardware. so for edge to work really well, especially in the telecom space, you need a lot of these infrastructures in place.
[00:43:43] And all of these changes in place and 5g is driving these changes. So this is what I see the value of 5g for edge is because it accelerated a lot of these changes on the legacy networks, so that it allows us to start running workloads where we need them to run on the network in an [00:44:00] efficient and a clever way.
[00:44:02] Matt: [00:44:02] Yeah, I think I agree with your theory that, That a lot of it is marketing. That's really interesting as a marketer. I never really gave it that much thought. because there isn't a lot of concentration and billions of dollars a and you know, people are searching for the business models. at the same time, you're, you know, there's some underlying technologies, you know, container orchestration, Kubernetes, things like that have made a lot of these things more increasingly, Not easier to do, but certainly from the under structure for doing these things and use your ways is coming about.
[00:44:31] And the growth of AI is giving us something to do with this, all the sensor data, which, you know, I think, you know, five years ago, what would you have done with it? Right. You know, some, you know, some big data stack that would run overnight. so, so that is really interesting. The marketing element was really a catalyst.
[00:44:49] So if you. If you can help us sort of cut through the hype a little bit. w what do you, what are you seeing? I mean, when you're looking at your business evolving, and you look at the [00:45:00] edge infrastructure and the whole ecosystem, because it really is going to take all of us working in concert to make this happen.
[00:45:06]w what do you, what are the milestones you're looking for looking out, you know, the next three years, what's most important to you?
[00:45:13]Mahdi: [00:45:13] so I think the first milestone that we're looking at from a, from an infrastructure perspective is, the adoption of, you know, virtual run. Because this will allow us to actually go into the far edge where we need to go, where we want to go, where we can show the value eventually of what edges
[00:45:31] Matt: [00:45:31] And this that's your local cloud, essentially.
[00:45:33] Mahdi: [00:45:33] our local cloud.
[00:45:34] Yeah.
[00:45:34] Matt: [00:45:34] Okay. Got it.
[00:45:35]Mahdi: [00:45:35] and that's a milestone we're working towards and, you know, we have plans for and how we're going to do it and how our software will interact with that and how the software will interact with the network and so forth. And that's a major milestone. the central cloud, we've already done a lot of work on that, and it's pretty obvious of how it's going to happen.
[00:45:51]it's just a matter of time in that space. The second milestone. And I think it's going to be the biggest trend in edge soon is an intelligence, [00:46:00] like how do you make these platforms more intelligent? Because in, in public clouds today, where do you have all the servers are fairly unified.
[00:46:09] They run the same type of platform on top and. Yes, there's a lot of intelligence there. The clouds are building a lot of type products that are really intelligent, but when you go to this distributed infrastructure over this fast geography, the game is different. You know, and how do you interact with networks?
[00:46:26] How does the network, like, how do we route things, how do you basically move workloads from one place to the next based on demand, based on, use of traffic and that's a really exciting challenge to solve and to work on. and I think that would be the biggest driver and it's a bit, it's going to be a big trend for a lot of players moving forward and edge as well.
[00:46:45] And for us to fully crack that, Is going to be, the next milestone for us in the next two years that we're working towards.
[00:46:54] Matt: [00:46:54] Using any open source technologies, to develop your cloud.
[00:46:58] Mahdi: [00:46:58] Well, we use Kubernetes. [00:47:00] Yes,
[00:47:00] Matt: [00:47:00] Okay. Okay. and does that mean that, if I'm a Kubernetes developer developing for the Ori cloud were very familiar to me.
[00:47:08] Mahdi: [00:47:08] for sure. I mean, if you're familiar with the Kubernetes API APIs, then you can develop against them and you can run your workloads against them.
[00:47:16] Matt: [00:47:16] And are you doing things like building a custom scheduler that introduces the kind of, kind of quality of service, or SLA
[00:47:22] Mahdi: [00:47:22] Yeah, I mean, underneath and around keeping that is there's a lot of custom, part of stop word building, even within the Kubernetes framework. So internally we use Kubernetes API APIs, And basically we're building different custom things inside it, but we're using Kubernetes as a plan as a framework, to allow us to give the seamless experience for the developers, but also it's, I mean, it's a great technology as well.
[00:47:46] So, it helped us a lot and it allows us to move quicker.
[00:47:52] Matt: [00:47:52] Well, I also think it starts to answer the question I asked a little while ago, which was if I want to be an edge native developer, where do I go? And [00:48:00] I certainly think, you know, the Linux foundation with the combination of the CNCF and some of the Kubernetes edge. Working groups and some just the core stuff.
[00:48:07]but also, you know, the LFS that there's a, there is a growing, developer community that at some point, I think will start calling itself an edge native to offer community. They haven't quite adopted that, but I think they're thinking that way.
[00:48:19] Mahdi: [00:48:19] I mean, w we're trying to actually get people to stop calling themselves that because that's the only way you can start getting people to ask questions, because how do you get things to grow? you create something that will make people ask questions, because people say, Oh, well, what's an edge native developer community.
[00:48:34] And then we'll go and research it and they'll find about it. Or what's an edge native developer in general. And this is where like, ah, I can relate to that. Maybe I can start building to that. And I think it's a really important thing to do as an edge community is to actually build that name so that people ask questions.
[00:48:49] What is it? And by asking questions, they'll adopt it eventually. we actually are hosting our first developer event, in November, we call it edge hog [00:49:00] day.
[00:49:00] Matt: [00:49:00] and talk. I love it.
[00:49:01] Mahdi: [00:49:01] Yeah. So our mascot is called edge hog actually. it's a hedgehog, as you know, and, we've invited a lot of the current customers that we're talking, that we're working with to come and actually talk at the event about their edge experience.
[00:49:14] So we're starting to push and try and build the edge native developer community in this space.
[00:49:20] Matt: [00:49:20] Yeah. Well, I won't ask you to commit to it live on an interview, but, I would love to have you write a, an opinion piece on edge native development for the sheer state of the edge book. Cause I think it's a really interesting topic and you've thought about a lot and speak very articulately about it.
[00:49:35] Yeah. So, let's, I want to wrap up a little bit, but I have a couple of last questions. so the first one is what's the origin of the name? Ori industries, kind of a cool name. What's the origin.
[00:49:45] Mahdi: [00:49:45] So when we were thinking about the name, We kept thinking about our platform as this biological fabric, like DNA that can actually talk to different types of hardware and compute and build this [00:50:00] common language between all of these different components, across the globe. So, and then while I was researching inside biology and DNA, the origin of replication and DNA is actually called Ori.
[00:50:13]Matt: [00:50:13] it also explains the, you know, the nod to biology and some of your product names.
[00:50:17] Mahdi: [00:50:17] exactly. Exactly. Yeah.
[00:50:18] Matt: [00:50:18] Yeah. That's really neat. So if, if people want to go to this developer conference, where can they find out more about it?
[00:50:24]Mahdi: [00:50:24] so we have a website that is, launching, next week at edgehogday.com.
[00:50:29] Matt: [00:50:29] Okay.
[00:50:30] Mahdi: [00:50:30] If you go to edgehogday.com, you can register and you can come and listen to us. You can listen to our telco partners and you can listen to some of the developers that are currently developing on our platform.
[00:50:41] Matt: [00:50:41] No, that's great. And if people want to find out more about Ori, wherever they go.
[00:50:44] Mahdi: [00:50:44] Just go to our website ori.co
[00:50:47] Matt: [00:50:47] awesome. And Mahdi, you personally, if people want to find you online, where do you hang out?
[00:50:53]Mahdi: [00:50:53] I'm on, Twitter. So it's Mahdi, M a H D I Y my name, and, you can go on [00:51:00] LinkedIn as well. So if you search my name, you can find me on LinkedIn for sure.
[00:51:05] Matt: [00:51:05] Yeah, Mahdi, this has been a terrific interview. I really enjoyed spending this hour with you. I love your pan optic vision for your company. it's so much fun to see an unbounded entrepreneur. you know, you have to be a little crazy. to succeed.
[00:51:19] Mahdi: [00:51:19] To go into agile. I think you have to be a little crazy, to be honest.
[00:51:22] Matt: [00:51:22] yeah. Yeah. You're a good crazy though. so thank you very much for spending time with us.
[00:51:26] Mahdi: [00:51:26] Thank you. I really enjoyed this. Thank you.
[00:51:28]