From Julliard to Bare Metal with Zac Smith

Avi Freedman: Hi, and welcome to this edition of Network AF. We have Zach Smith. Zach has started multiple companies in networking, and we're going to talk about how to get into networking, how we learn open source in networking, cloud versus traditional works, and sustainability as well. Hi, and welcome to Network AF. I'd like to welcome my friend and fellow networker, Zach Smith. Zach actually is responsible for helping me found Kentik indirectly, in that he introduced me to one of my co- founders, Justin Beagle, when Zach thought that I was still in hosting colo cloud, and Justin was looking to stay in hosting and cloud and introduced us. Justin discovered I was not doing that anymore, and was like," What is this analytics stuff? And-"

Zach Smith: What is this amazing network stuff where I don't have to take all the server calls?

Avi Freedman: That's right, and I'm not in the middle of the path if things die, and companies don't tell me to get all my servers and move them from the building because Google has just bought it. Well, thanks again, Zach, and could you tell us a little bit about who you are and what you're up to, what you do now?

Zach Smith: Sure. Awesome. Hey, Avi, it's so good. I'm a little flattered to be included in the network guys thing with you, because I wouldn't normally call myself that. But a little bit about me. So yeah, my name is Zach Smith, I live in New York. Been working in and out of the internet plumbing business for about 20- some years. I landed into it like most people, totally on accident. Had a formal training as a classical musician and then needed a job, like all well trained classical musicians, and ended up getting into the web hosting business in 2001. After that had to learn around networks and build some things because it wasn't so easy back then. You had to hike up the data center stairs, both ways, up hills or whatever it was, cable your own cat 5, so yeah, we had to build the stuff which is-

Avi Freedman: I crosstalk. I believe cat 5 should be excreted from the back of the machine. I don't understand, that's all.

Zach Smith: Right. I did a couple punch downs and polished some fiber here and there, but not very well. So I worked learning the stuff and figuring it out along the way, building some networks and peering or at least around people who were doing so, and kind of couldn't get out. Spent the last six years building a startup called Packet where we automated single tenant hardware around the world. And in March 2020, that business was acquired by EquineX, so now I work at EquineX where I run the EquineX Metal business. And what we do is basically help our customers access the global reach of EquineX, so 248 centers all across the world, with less friction than what you do with colo. And so we're building an automated colo product attached to our interconnection capabilities, so that's what I do these days.

Avi Freedman: Awesome. I'm a customer of EquineX, as are most of my customers.

Zach Smith: Thank you very much.

Avi Freedman: And our whole environment lives in Packet, so thank you very much.

Zach Smith: Love it.

Avi Freedman: So how did you get started? You talk about that, what was the transition from music to networking, which I know other people who have done also, but this was...

Zach Smith: We do an all hands every Monday, and it's increasingly number of people, and we started to put on during the pandemic tiny fortress concerts. So the EquineX thing is like a fortress, right?

Avi Freedman: Uh-huh(affirmative).

Zach Smith: So instead of the tiny desk concerts, we do tiny fortress concerts. And like every week or every other week we highlight a musician in our ranks, because turns out there's tons of musicians who work in networks and whatever. I have this thesis that most people in the plumbing of the internet either were the stage hands in musical theater in high school, or like-

Avi Freedman: My brother.

Zach Smith: Yeah, or-

Avi Freedman: My brother who is crosstalk.

Zach Smith: ...were in the band.

Avi Freedman: Yeah. Tech crew.

Zach Smith: One or the other. My story was not that glamorous when I was working here in New York City to pay my way through college. I spent a bunch of time basically doing PC repair in 5th Avenue apartments for rich old ladies. And so I figured out that I could make more money getting people connected to the internet with AOL discs and figuring out why their printers didn't work than working at the box office at Juilliard, handing out tickets for minimum wage. So that was my intro into computers. I'd always liked computers. Some of my friends when I was younger were really into gaming, so I was the guy who lugged around the PC and they were the ones who did all the gaming, but it's cool. So I had been comfortable with technology in a little bit of way, but never really thought about it as a career. I was studying for music. So after Juilliard, I needed a job and I ended up working a night shift at a bank doing PowerPoint. 6: 00 PM to 4: 00 AM working on PowerPoint decks for bankers who actually never showed up to work. They were out with customers and I was sitting there. You couldn't leave, because they might come back and they needed to have you update all the prices in the pitch deck before the next morning. And so I had a ton of time and access to the internet. And so I started thinking. This was just before September 11th, and they were paying ungodly amount of money for us to sit there and do nothing. And I was like," This is not going to last." So I had to figure out a real job or some sort of thing.

Avi Freedman: Active.

Zach Smith: Yeah, something. I was like,"I got to get ahead of this." And so I said I'll start a business, because nobody would hire a Juilliard grad into a real career. And so I called a friend of mine, his name's John LeRoux and he was a family friend, the only person I knew who had ever started a business or whatnot, and my parents were both in trade businesses or whatnot. And he had started a CLAC called Pac- West Telecomm back in the mid'90s.

Avi Freedman: Oh, I remember.

Zach Smith: Yeah, and providing dial- up PAC- call and things like that. And I said," Hey, John, I'm thinking of getting into business. Should I start a phone company?" And he's like," Absolutely not. Do not start a phone company." But he said do start something with recurring revenue, because if you charge people, sell them once and then you don't mess up, just treat them well, they'll usually pay you every month and that's a good business model. And so I basically decided I was going to sell web hosting to musicians.$ 20 homepages, put them on the internet. Everybody needed a business card on the internet at the time. And so I went to webhostingtalk. com and posted at 2: 00 AM in the morning on somebody who could sell me a Lunix server.

Avi Freedman: A Lunix?

Zach Smith: A Lunix server because I didn't know... I'd heard the word, but I didn't know how to spell it.

Avi Freedman: That's great. I need a Lunix distribution.

Zach Smith: I needed a Linux distribution, and lo and behold Raj Dutt messaged me back from this hosting company up in Troy, New York called Voxel, and sold me a server basically, because I had some time and money and he had servers and knew a thing or two about Linux, it turns out. And so that's what got me in, and a couple... I remember it was a couple months later I had sold... It turns out I was pretty good at selling, so I had sold a couple 100$ 20 a month web hosting accounts to musicians. I was kind of like the hookup for cheap websites on the internet for musicians in New York. And so I put on a suit and I drove up to Troy, New York to go meet-

Avi Freedman: To see Raj, okay.

Zach Smith: ...to see Raj. I was going to go and see what this big company was about and see if I could really do some business and whatever. And I show up there and it's Raj and Matt and a couple of his roommates. Raj was late by a while, and they were just college kids who were either at RPI or had dropped out of RPI, couldn't really tell. And so we decided to go in to business together. Took a while, but I became partners and I did a lot of the operations and success selling side, and Raj is running all the product and making all the good stuff, and we created a pretty cool cloud company in CDN, and that's where we really got into networks because turns out at the time, IP transit was expensive.

Avi Freedman: Oh yeah.

Zach Smith: And the question was always like," Wait, I need to put a lot of servers at 25 Broadway Telehouse, but they don't have a lot of power, so where am I going to put the servers that can be near the network?" And it turns out that was really expensive, so we needed to put the servers someplace different from the network, and we had to start getting into some DWDM and building a little Metro Ring and figuring out how we could back haul to the right places. So that's how the deep dark hole began.

Avi Freedman: That's awesome. You make me remember when I used to go to computer shows and buy computer parts and put them together for people. And I remember a moment of crystal clarity when I told the first customer, because of course, then they all wanted help with their disk is full, and they can't use AOL or whatever.

Zach Smith: Because you were the computer guy.

Avi Freedman: I used to go to Micro Center and buy a compact with a three year warranty, that'll be cheaper and you'll be happier. I'm like," No, no, no, we need to buy it from you." I remember when they bought my ISB and the company that bought the company that bought the company finally shut down the shell accounts. We were like," You need to do a shell business." I'm like,"Go get a VPS. You'll be much happier." And-

Zach Smith: It'd be so much easier.

Avi Freedman: Yes. Like you, I discovered the joy of recurring revenue. This is the first SaaS Kentik, but recurring revenue, as you said, if you do a good job, don't view customers as a nuisance. I sort of laughed at this whole network as a service thing now, which I'm sure you have a product for.

Zach Smith: Right. Wasn't it always a service?

Avi Freedman: Exactly, but it's the modern approach of," Hey, I have a service," that means I invest in the customer, and things go well.

Zach Smith: Well, maybe it borns from the whole, when you had to be my customer, because I was the only network.

Avi Freedman: Right.

Zach Smith: That it wasn't really a service, it was just like a toll.

Avi Freedman: Right. Yeah. Well, yes. I mean, in monopoly subsidies and all that. And it's funny because I had Elliot Noss on, who's at Tucows, and I remember almost 10 years ago, he is like,"Avi, do you want to help me do fiber ISP?" I'm like,"ISP? Who wants to do that anymore?" And he's like," No, no, no, it's..." And it is a mitzvah, it really is to help.

Zach Smith: Bang.

Avi Freedman: But yeah, it is a thing. And it's funny because I was Akamai, I remember hearing about Voxel. You wound up getting a number of really good people. I remember when TCAP was involved with you, I'm like," What, they're building a CDN? Who builds a CDN and storage?" We were talking about object storage and all this stuff, proto cloud. And of course the testament is you look at what's come out of that. So Grafana, Raj as you mentioned, NS1, Packet. I consider myself an honorary Voxel martyr crosstalk-

Zach Smith: Oh, Avi, that's so cool.

Avi Freedman: crosstalk. Was that a challenge doing so many different things in a company? How many people was Voxel if you can share again.

Zach Smith: I think at our peak when we sold the Internap in 2011, I think we were 90 people or something, and we had a lot of products. And it was interesting because I've been thinking about this recently because one of the things that Raj and I as founders or as partners were very different. I always consider us to be like oil and water. And I thought that was a negative, actually it was a super positive looking back at rears, with more space and time between you and more maturity, because we were 20- something year old punks at the time. We had very different strings. And one of the things that I was always focused on, I was kind of... I'm the kid who wore my socks to bed just so I could get up and put my shoes on. I didn't want to be late for school. That's a little bit different with Raj, who's more focused on the ideas and the creative. He frankly was a very very creative individual. And I was very like," How are we going to solve this problem? Where's the process for that?" And so I was always really focused on the now in our business. How could we make money with this? How could we make sure to just stick to this thing? And Raj was always forward thinking. Frankly, in the rears when I look back he was already thinking well into how a cloud was going to work in 2003 and'04 when we started hacking with Zen, when we had no business hacking on Zen and hiring our won developer.

Avi Freedman: And I remember crosstalk was object storage. I'm like," Why is an ISP thinking about object storage? That's awesome. I love talking about object storage."

Zach Smith: Totally. We were investing in swift and committing on that. It was way too early for our own good. We didn't have resources for that, but I think Raj really saw where this would go, but we just had bigger eyes than our stomach would allow, frankly. And that pushed us, but it also attracted a lot of really great people to the mission of utility scale computing and highly programmatic access to infrastructure primitives. We were pushing that thought process. I mean, execution was I think, great due to some individuals, but from a business perspective, I think we maybe didn't capitalize on all of that, but we certainly pushed the limit on where we were thinking. I also would say that we just didn't have a lot, until maybe 2007 or'08 we didn't have too much to lose. As I was always," Keep reinventing yourself and see what sticks." And it was a pretty exciting time of early cloud computing, because, how many hosting companies were around? It was very fragmented. There was just so much going on. It's obviously consolidated to a large degree I think at this point in the internet infrastructure space, but individual regional data centers and peering exchanges were popping up. It was still a little wild west. So you could afford to go out there and throw the football and see what happened.

Avi Freedman: Yeah. And I was through probably some part of my own, I was communicating too technically and not at enough at the business level. I was at Akamai and for about eight years tried to get us to build something, which is now a thing, native edge computing.

Zach Smith: Right.

Avi Freedman: Because we had a java- based edge computing, but it would've been a different margin profile, a different capital profile. In the earlier days things crosstalk-

Zach Smith: Different customer profile in a lot of ways too, is develop a customer.

Avi Freedman: Yes. But a lot of our customers ran their origin. So it's like people, what can be abstract? If you look at what Akamai did with early provisioning system, commit rollback of systems, auditing, logging, message... it looks like a lot of those systems if you have a planetary skill computer.

Zach Smith: Yeah. It's big distributed system.

Avi Freedman: Yes. But could not get that going. So it was fun to see that kind of innovation. So the internet community is a little different maybe than some, there can be some hostility, but there's also competition. What was it like when you sort of got in? Did you have mentors? I don't know whether it was from the investment banking days or in the company outside the company. How'd you get in?

Zach Smith: Well, we had a really interesting setup. First and foremost, there was big culture of open source at Voxel, which was how to contribute input within the community. We were a big committer to Debian pushing out as many things that we could into the open. And our first, one of the projects that we struggled with was an open source billing platform called CBMS, which then became a proprietary SaaS platform called Ubersmith. And the reason why is because we couldn't afford to continue to invest the monetization angles of SaaS hadn't really occurred yet. So we were struggling with how to build this, but it was very community driven. So for years, I mean for the entire time that we ran Voxel and we also had this subsidiary called Ubersmith, which armed and provided billing platforms and device automation to basically all of our competitors. And so we were running the back end of hundreds of different hosting and cloud companies as well as building our own. And it was an interesting line to straddle for us. But what I always found in the hosting community and in cloud in general, networking I think more broadly is, it is a community. And that's what I actually align between the early open source as a thing. And you can see it in other things. Network for sure. Now people get together. Internet only works if people are figuring it out together, and there's trust there and you got to build those relationships. I mean, one of my first things in networking was establishing settlement free peering. I realize that that was actually built on trust with people." Are we going to basically send traffic to each other's to networks, is that how that's going to work and maintain capacity together? And that's going to be good for both of us." And so doing things like that, I think you can see it in new, maybe not as much in cloud, which is more of, I'm going to call it a verticalized proprietary approach to infrastructure. And part of, one of the reasons why I started Packet was to help try and normalize that field a little bit. But I see it also weirdly enough in the one that reminds me of my early days in networking and hosting is actually blockchain. That is also really this decentralized, distributed community treating thing, where rising tide reaches all boats, make the tent bigger, versus capitalize on the thing just for yourself. And I found a lot of community and partnership throughout that. It wasn't always shared with other maybe hosting companies and whatnot in earlier days. But I found through open source alignment and then through networking that those were very very inclusive communities that always wanted to help talk about it. They're passionate. They had robust communities and knew that they needed to figure out how to work. Otherwise, the whole thing didn't work. And I think that's been one of the superpowers of infrastructure businesses, have been that community- driven approach. And I'm hopeful that in things like SaaS applications and cloud that that can also be a thing. But for me, it was always great because I never... You had to get over, and this is something maybe I learned at Juilliard, but you have to be okay being the stupidest person in the room. I went to Juilliard and I was," There are two out of a 100 people who auditioned for the double base thing got in." So I was like," Not bad," but I was absolutely the worst at the school. Let's be clear. There were virtuosos and I was making it along. So you have to be willing to be humble and eat that. And when I first got into networks and hosting, I don't know anything, but as long as you're willing to ask more questions, I've found that people are really interested to share and help, which is pretty cool.

Avi Freedman: I love saying I don't know yet. And then one of the things I try to advise people to do, not just earlier in career or younger is, when something was confusing, often you don't want to admit that it was confusing because it's like," Oh, well, everyone understands." But it probably means that it should have been explained better. When I got into networking, it was all this Cisco mumbly book about the NLRI, blah, blah, blah...

Zach Smith: Social command that you had to know.

Avi Freedman: I was talking with someone who works at VMware and I told them I had to stop reading the Yellow- Bricks blog, because I know what the things are, but it's like a stretched V stand, blah, blah, blah." But the open source is sort of the antithesis in some ways of the vendor ecosystem of everything, is you have to fall into the certification hole and do that. I guess question for you, is that doing well by doing good? Have you seen a positive ROI on that at Voxel and then at Packet? Because I saw you were very aggressive at Packet giving resources to opensource and recently opensourcing, not that recently. COVID creates this air gap. I remember in person is when you did the announcement at Reinvent of opensourcing your provisioner.

Zach Smith: Yeah. It served me well. I think that overall the concept of creating bigger ecosystems and more opportunity generally, like I said, a bigger tent, has proven to be pretty good. And I think, especially in technology where it's all about opportunity, there's more things around the corner. We just need to encourage those to have happen. And so being a part of widening the tent or broadening that has always been a good strategy for me. It's also part of broader mission. If your goal in your job is to," I want to be the number one this that provides all the things for that." Okay, that's a goal. There's also a broader goal at least. I've attached myself too is, we want to enable this for the world. And however that happens, hopefully we're going to play a part in it. But we still want to enable this for the world. Like when we start a Packet, the idea was that software was becoming more portable, becoming more distributed and more, I'm going to call it open. And yet infrastructure was actually becoming more closed. It was harder to use and less accessible. And I thought back in my early days as a career and how we started Voxel, we went to Alex Rubenstein at NAC and he gave us a half rack and a borrowed switch and showed us how to be on the internet and helped out. That would not be very possible today for somebody to just walk in and get a half rack of colo from somebody who helped them get going on the internet. And so in a weird dichotomy, you had this instant easy consumption- based access to infrastructure and backbones and thing being majigger, but you couldn't actually innovate on it at your own pace because you weren't allowed to go play around on a global backbone. You could just use it.

Avi Freedman: You could just connect and break things. But now people don't like it if you break the internet.

Zach Smith: Well, it turns out there's real things going on, on it. I'm sure we'll get to that later. So one of the reasons why we had started Packet was this idea that we wanted to enable easy and better access to technology infrastructure that could be paired with software so that magical things could happen for the world. Sustainability could be achieved through better use of our resources. Innovation could happen that was maybe inaccessible in one or the other on its own. And I struggled for many years, we tried and we wrote a blog post back in 2015 about how we, I was like," OpenStack is ready. We're going to do it. We're going to get in there with money and resources. We're going to fix bare metal provisioning in OpenStack." And it turned out, no, we weren't. So...

Avi Freedman: Too many choices. OpenStack had too many choices.

Zach Smith: What do we?

Avi Freedman: crosstalk. Sometimes you need a little bit of in the middle, like, here's the things, here's the way to do it. Not five choices for everything.

Zach Smith: And so we had thought open was the way and it was, it was just the wrong place to do it at the time. And so we ended up creating a proprietary set for automating physical hardware, how to make physical hardware accessible to developers. That was our mission. So we ended up creating this. And then we really wanted to open source it early on in 2015, actually. But we struggled with," Well, how do we do that when the community is so niche?" There wasn't that many people who cared about it. And how could we? The worst open source projects are the ones that die alone on the internet. And so how could we do this when we couldn't find the community. And so it took us a while to build a community that we knew cared deeply about fundamental infrastructure, just like we did. And then eventually we were able to find the resources and support bringing that out. And now it's pretty exciting to see what's as part of the CNCF. We think fundamental infrastructure paired with portable software is an incredible boon to not only our business, but tons of others. And I think we're firmly on the broader, bigger tent is going to equal more awesome things at this point.

Avi Freedman: So if the open source isn't decor, what we've seen or I'd say struggled with, is, it actually takes more resources to have that community involvement, because of course, then you get everyone's opinion. When we started people Netaxs-

Zach Smith: Right.

Avi Freedman: ...people were like," Oh my God, you have an awesome colo store database. And it's much more efficient than Lastic. Can we use it?" It's like, well... But then ironically you have the snowflake problem, not that snowflake, but everyone wants to do something different problem.

Zach Smith: Of course. It's hard managing and building democracies. It's a lot of work. It's not always clean, and so the benevolent dictator side of thing might work better. And that's where we ended up, which was until we had the resources in a eight- person developer relations team and dedicated engineers whose whole job was to support the community, we couldn't do it and show up with the right integrity, frankly.

Avi Freedman: Okay. That makes sense. So is there something that, as you say, looking back that you think has, you've done, you and your partners have done? I'm also old, so sometimes I say partner, instead of co- founder, have done that has enabled people to...

Zach Smith: Avi, you're a tech startup forever. It's all good, buddy.

Avi Freedman: I will always be. So, I have the ideas, we need to... Again, it's focus. You need to do this, you have focus because there's so many things to do. Is there something that you've done, or a set of things that has enabled so many of the people that have sort of worked at Voxel and even at Packet to be the next generation, to start? Because I sort of think that one way to judge how you're doing in your career is the people that you've worked with, how are they doing? What are you enabling?

Zach Smith: It's a nice reflection. I don't know if I have all the answers. I mean, first off, I love not only creating product, I love creating outcomes for people. I think it's really exciting. And one of the true joys I've had in my career is being able to do that. I believe that in infrastructure, digital infrastructure, cloud and networks, it's one of those areas that is still apprenticeship driven. You don't go to get the master's degree in cloud computing and then start a cloud computing business. You show up and work next to the person who's running the network, and then you start figuring out how to run the global backbone, and then you know that's how it goes. To me, that's why I called it internet plumbers early on, because I actually believe that it's a trade that is super lucrative and necessary and exciting and whatnot. So if you have desire and interest and passion, there's plenty of ways to catch up. I always tell new people coming into the industry," Don't worry, it's all changing in the next one or two who years anyways. So you're not that out of date. So it's all good." And I think being really open to... Now we call it, at EquineX, New to Career, but not treating new to career as," Well, this is your first job, you should know the things." More of," let's get you on the journey. And it's our mandate, my mandate, my accountability and responsibility to mentor you, to apprentice you, into this business." And that will be a good thing for us because number one, probably will be a great employee, loyal and learn. And if you're passionate about it, there's all kinds of great stuff that we all need figured out. And number two, that will create a career path for somebody. At Voxel Raj had a special quality of really identifying people who were driven to a mission, and helping take in anybody who frankly had that desire from a mission and find the right work for them to do. Some of the reasons why we created whole products were because," Wow, this person really is great and wants to work on... is passionate about what we're doing, but we don't have a position. Oh, let's create something for he or she." And the other was just providing that training ground and being really open to role migration or accountability." Oh, you've only been here six months, but you're doing a great job, and now you're the boss. Great. Let's do it. You can do it. You're just getting started? Is all good." And I think that having that and supporting people has just allowed that to occur. And luckily we're in an industry that has been growing a lot over the past... if you like it continues to grow.

Avi Freedman: Probably will than what happens in the greater market.

Zach Smith: Yeah. And so not only was it, use the phrase, doing well by doing good. That was selfishly like, what were we going to do? Go compete for the talent that didn't exist or help grow new talent, create new awesome leaders and people. So I think that's just been a mission that I've always had both selfishly, but for the community, because it helped me. I just came into this industry as a bass player, I didn't know anything about Lunix, and yet it's been really well for me because other people said," Hey, yeah. That's okay," said Raj," Let me show you how to do this." And Adam Rothschild would always answer my questions. And then Alex would give us a colo rack and say," It's cool, let me show you." And those things people gave and opened doors from a technical standpoint. And then from a business standpoint, just had a lot of great breaks along the way. Frankly, I always consider startups to be messy and you can't figure out timing sometimes. It's just like the world aligns, but you can increase serendipity. And so creating-

Avi Freedman: You put yourself a position to take maximum advantage of getting lucky, but you do need to get lucky sometimes.

Zach Smith: ...Absolutely, need to get lucky. So sometimes the people you just meet along the way, as long as you're receptive can really just change your trajectory. And for me, I've had several of those from Raoul Martin that came into Voxel in 2009 and we were out of money. We were growing so fast that we couldn't afford to buy anything, and nobody was giving any debt at the time. We had this super weird position of," Oh my God, we're growing really fast, but we can't grow because we have no money," because we were bootstrapped and had no capital base. And Raoul was just such a great partner. He helped us raise the capital. He helped institute a lot of the things I live by today and running businesses that were not... I was always over of complicating it. And he's," Listen, it's not that complicated. You have the right people in the right seat doing the right job, and they're probably going to do amazing things for you, so focus on that." And," Hey, Zach, stop worrying too much about things you can't control and start figuring out what you're going to do about that today." So he gave me some really just incredible personal coaching that showed me how management and leadership and humility along the way could really create better outcomes. So that's been a pretty awesome ride, is to meet some of those people along the way.

Avi Freedman: So we talked about a few things that are related. Getting them young or at least early, it could be someone who's 50, but new into the industry, but especially getting people early in career is a much more diverse pool of candidates than if you say people that are later in career. So if you're trying to build-

Zach Smith: Especially in our industry.

Avi Freedman: ...Yeah. I mean, the peering world is more diverse than say the core internet engineering world. At PacketFabric, Anna talks about trying to sell to computer science people that, as you said it, internet is distributed system. If you're a math person or if you're a computer science person, wouldn't you like to study the biggest distributed system, that's the most critical. At MyNet access, Alex had the other NetAccess, NAC net. At MyNet access, we looked for the bright shiny eyes. But it's actually a positive thing to say," Hey, I don't know. Can you show me? Or, I'm interested in systems or web hosting or networking or-

Zach Smith: Totally. Oh my God, if somebody shows up and they raise their hand saying," I am super interested, don't know anything about this, but I really want to learn," you're like," Awesome."

Avi Freedman: I almost feel like we need to... sometimes we need to maybe anti- train people that in college it's sort of like," Oh, I'm going to get a J- O- B and they're going to tell me what to do." And that isn't necessarily encouraged. So I sort of feel sometimes when we get people that come through, I'll say," The system." I have all, but two classes of an undergraduate degree. So I went through the system. I have one semester grad school actually at Stony Brook, which Gail did not like living out there. But my ISP like you had already started and it was clear was either going to make money or make me very bankrupt because I could buy an infinite number of modems, but I didn't have the capital for it. So what's the trade off between, or I guess, let me go back, mentorship and the ways in which this is similar to being an artisan in the 1200s, that's good, but is that trickier with COVID? Does it sort of deny us and make it hard? I think sometimes one of the challenges is how do we help formalize some of it in a world where we don't get to pattern by hanging out with each other as much as we did. Is that you say net positive or are there things that you think about that we could improve with that trades person and apprentice model.

Zach Smith: I think it's a net positive. I think that the concept of like," Ooh, we'll hang out at the water cooler," was already pretty selective. And so, if you had a badge to work in 111 Eighth, and then you were already there, then we could hang out at the water cooler. So I think that flattening in terms of remote has been awesome. I spent a bunch of my time... I'm the chair of an operating board of a nonprofit called Pursuit, which is based in Queens, New York, and helps bring underrepresented communities into the technology ecosystem. And we mainly focus on full stack developers and whatnot. And what's been fascinating about that journey is our average person who applies for our program is making$ 18,000 a year, usually supporting two to three generations. And after a one year program that we put through nights and weekends, and all the other times where they can actually participate, including getting internet access and providing computing resources and finding quiet places to work and do your studying, income average across the cohorts is$ 88,000 a year, which is pretty incredible because it doesn't change that person's life, it changes everybody around them. And creates tons of possibilities for," Oh, I didn't know. Well, she did that. I guess I could do that." And that starts to happen and occur. And we've graduated several thousand people into tech jobs at this point doing this. What I love about that is, we always had a barrier frankly with pursuit. And I've seen it as I've entered into a larger company. I never had it at Packet where we had a remote first. We don't care if you have a degree, like felony background check on drug misdemeanor drug thing, not an issue for us. We had very, what I'm going to call progressive hiring standard. Well, we also had to, because we were looking everywhere in the world for talent, with certain set of budget that we could afford and certain things. And so you were always willing to reinvest in a globally with lots of, I'm going to call it the more thin rules, obviously basic background checks and things like that. And what I found at larger companies is there are systemic things put into the hiring processes that really go against this. And so we've worked to change that here at EquineX, whether it is where your home out of, you had to be out of certain offices, whether it was minimum degree requirements or skill sets that naturally went against new to career or anybody with a non- traditional path in terms of education or whatever. Like for an entry level job, must have at least six years of experience in the career. I'm like," What, how's that going to work?" And so figuring out how to change some of those things and then build. By the way, they aren't incredibly successful talent acquisition strategies, long term, because you're competing with a very very aggressive technology landscape these days as you know for talent, and so you end up fighting over the same things. And so what I've tried to do is help, and I think it is an opportunity, I call it being long term greedy, which is, is it easier to hire the person that already knows how to do the things? I guess so, maybe, if you can find the person, if the budget works, if they stick, if the culture fits, all these other things. Yes, it might take a little longer to invest and mentor somebody. Maybe it costs you more resources, but actually long term greed, like average it out in like a startup is where I focused. Because averaging out over a few years, actually think that the invest in people and grow them strategy actually has a better ROI in almost all cases from what I've seen, at least in my career. And one of the areas that I worked in Packet was venture backed. I know we shared some similar investors and whatnot, was getting buy- in from my investors who immediately turned to," Okay, when do you turn on the hiring and the recruiting because we need to go fast, because you just got your series B?" I'm like," Well, you know what I really want to do? I want to go and invest in a mentorship program and go longer and bring new people because that's going to be a talents strategy for me over the next 24 to 36 months." And they're like,"That's way too long. You need to go get people now." And so trying to get the buy- in from venture capital, I think is also really important. And get the social accountability side for VCs to say," Hey, when we do your series B, we are going to make sure that we support and fund a mentorship program, an apprenticeship program, because we know that that's a hiring strategy that will be sustainable in our inclusivity goals as well as getting talent.

Avi Freedman: You should blog about that. I think people understand in the VC world the compounding nature of SaaS as the inaudible writes about, the investments we make and the virality of it. But often I've seen... Not my investors. I have great investors, but investors where I've said," Hey, I see this company shutting down. We can help place those people."" Not my problem, go talk to the whatever interim CEO I hired to do the company. I'm onto the next thing." But really it helps the venture ecosystem and even other companies in tech.

Zach Smith: I would say, you probably see it. But I think one of the biggest risks to our industry in general is the lack of talent. We need more talent, we need more diverse talent. We need a broader base of ideas of talent. That is all good for our industry. So I think supporting that in venture companies in Fortune 500 companies, like what I work for. And I'm really proud that EquineX has a strong inclusivity message in BIB. They've been extremely receptive and changed tons of things related to our hiring. Actually the pandemic and the remote work situation has actually forced a lot of those things because everybody's rethinking there talent strategies right now, because of global-

Avi Freedman: Yeah. When we started, investors were like... Pre- COVID we were two fists local in the Bay Area and three fists remote. And people were," Is that a conscious decision?" It's," No," as you said, we just need to hire people, and people will hire them. I think the industry struggles with, how do we do that? I'm told, I can't say osmotic learning, because those are two big words, but the apprenticeship model formalize it and...

Zach Smith: Yeah, I have some inspiration there. And for me, I think you have to commit to the model. By the way, I work at a company now that has 13,000 people across the globe. The concept of being in the same place at the same time is not an option. And so as a knowledge company, we have to find ways to be inclusive. For example, Zoom meetings are pretty cool. They're not inclusive when you have people all way around the world and asking people to get up at 4: 00 AM in the morning to join your all- hands. Well, I guess we should really think about how we make communication, knowledge sharing, participation to be inclusive regardless of time zone, or language barriers or whatnot. I'm inspired frankly by open source. My early days participating in Linux had three communication modalities. One, message boards, threaded contextual conversations that were asynchronous.

Avi Freedman: Usenet.

Zach Smith: Usenet. Asynchronous communication that have deep context that you can still go and Google and find the thought process behind very complex technical decisions in the Linux colo, because it's there. That might be useful right now, but somebody coming in who wants to add to that, didn't need to be there at the time. They can still get the context, which is so powerful. The second thing is then real- time collaboration; IRC, Slack, Teams, Zoom, whatever, it's real- time. We can meet. Wherever we are, we can have a communication. Time zone obviously being the challenge, but we can do that. And the third thing, which I think is what we really got to get back to ASAP were hallway track conferences. Get together, break bread, build emotional connections with people. And I think like if can mix those three things, not realize that it's just one, is asynchronous but deep amount of knowledge. Synchronous, but relatively hard for context or large amounts of volume sharing. And third is human connection. I think that's a strategy that can work across apprenticeship models too. You just have to be fo... You can't be," Oh, it was easy and they were sitting next to me." Well, it's not. And so we have to design around those things. And a lot of that will have to do... I notice things like onboarding new to career people its hard.

Avi Freedman: Oh yeah.

Zach Smith: Takes a lot of work. And I think back to, I used to, is it like GE or somebody who would do rotations? You would spend the first six months rotating throughout every part of the business?

Avi Freedman: Yeah.

Zach Smith: I started doing that-

Avi Freedman: crosstalk.

Zach Smith: ...and working pretty well at EquineX, we're starting to rotate people into the data center and then rotate them into sales, and rotate them into success. And yeah, it takes a little while for us to get through the rotation, like three months, but man, they come out with a lot of contextual knowledge and a ton of relationships.

Avi Freedman: And yeah, it's the relationships. At Akamai under a month, maybe it was a month, I don't know. After I started, we started network architecture inside engineering. And then in a month it's now a whole separate group. We're unifying the business strategy and the technology. In that month, now, it was bubble 1.0, we added 400 people in that month. The relationships that were built by having had that insight engineering and help all the way to this day, where my brother still has that job running the network group. And yeah, its still alive.

Zach Smith: So rotating through to meet those different people, which might take you one or two years, if you were just sitting in your siloed thing and reaching out and trying to create those connections, if it was a force rotation. So I think there were strategies. Let's put it that way. What would be super cool if, and I think that this is an opportunity for companies to work together, maybe in the networking space, maybe in cloud infrastructure, there can be blueprints. You don't have to invent all this stuff. There can be, please fork my repo on how to do an onboarding process that is inclusive? That would be cool." When we have spare time, come on, you and me on the weekend."

Avi Freedman: I'll give it to first round. First Round Capital runs a Quora- like network for... they call it the network, and they're trying to help with stuff like that.

Zach Smith: Oh, sweet.

Avi Freedman: I'll see if I can get you an honorary invite. They made the mistake of running contests on who can answer the most questions. So I like to win those.

Zach Smith: I read on Quora that somebody had answered 250, 000 questions on sequel all by himself. And that's the Guinness book of world records for most questions ever to answer on sequel.

Avi Freedman: That's a lot. crosstalk keeps inviting me to earn$ 500 a month by being an expert. But I've been off and off of it. But I still have people, our sales people go into companies. They're like," Oh yes, I remember reading the BGP thing, that you wrote in the nineties." And like," Okay, well, we got the meeting." There is a long tale to opensource and helping people and having people that have been in the ecosystem go.

Zach Smith: Totally.

Avi Freedman: I still think that we have to think about some of that apprenticeship, as you were saying, documenting across the industry. And I know inaudible is thinking about it and it's stuff. But for someone that wants to break in an open source you go, people can look at GitHub and you experience working in teams that frankly, just having a college degree, doesn't usually, even didn't really used to, but doesn't necessarily. Well, what can people stand out if they're interested in showing that they have the bright shiny eyes and they're interested? What makes someone interesting as...

Zach Smith: Obviously GitHub is just a great place because it's super transparent. When I interview people, much I don't do very often these days, but when I get to interview people and they say," I'm super passionate about this." I just look them up on GitHub. I'm like," Really, are you? Let's just go look."

Avi Freedman: Oh, do you even help with documentation?

Zach Smith: Exactly. There's tons of areas to participate and you don't have to have any technical knowledge. And if you're passionate about things and you show up with the green thumb and," I literally know nothing about this."" I know nothing about Go BGB," says networking group." But I would like to help clean up the issues list," or," I would like to help run the contributors meeting every Thursday," or," I would like to take care of the calling of something, please teach me." People will be," Oh my God, yes, we could use your help." And I think that some of those ways you have to learn different skills in this remote world now, because before maybe you could go to the meetups, which are a little harder to go to, or you could go to the conferences and hang out on the floor and get the stickers and start to talk with people and ask questions. You just have to distance different strategies now, but I think that those are all possibilities still. The world of SaaS is also cool because you can show up and talk with the people on chat and ask them questions and find out," Can I get a free trial to use this?" And people will generally say yes, and then you can go write a blog post on," Hey, here's what it was like to sign up and use Kentik for the first time. And here's what I learned." Wow.

Avi Freedman: Notice you.

Zach Smith: You would notice. And it would be," Hey." And so I think that being genuine and vulnerable in that regard for somebody new to career is actually great because most people will then really empathize with that and help. And so it means," Hey, I signed up for something because I'm interested in networking. And I used it and I had a lot of trouble because I don't know these things. And I asked these questions and here's what I learned." That's being open and that's being vulnerable. These are great qualities for new employees. And I think it's also a great way to build your authenticity, which is frankly, what I look for in new to career, is," Do you have the drive? Are you going to think outside the box to come up with, to satisfy your questions, and are you genuine? Do you want it? Is this important to you?" And if so, man, we can, we can impart lots of knowledge. That's not going to be a problem.

Avi Freedman: We look for passion especially early career in something. What's the thing? It doesn't have to be networking or technology.

Zach Smith: It could be customers. It could be-

Avi Freedman: Yeah. Or music, or-

Zach Smith: Document, I really care about Wiki Mako.

Avi Freedman: ...that thing, or poker or something, the thing which gets someone to," Oh yes, but..." have some opinions with humility, but have that passion and hopefully they can find that.

Zach Smith: I think there's a great op... Like in the old world of ISPs doing the tech support, customer service, I think it's just an awesome opportunity to start your career in customer success, where it's really about being with people and being interested in the product that you're representing and the customers and what they're trying to do with it.

Avi Freedman: And their life.

Zach Smith: There's so much opportunity in customer success right now because all of sales is transitioning from this sales led motion to success led where being deeply passionate about helping people use the thing is the game. And so that is just a great entry level, where you don't have to be in sales. You have to get into that, but you can be in the," Let me help people." And such a great way to learn and interact a dozen times a day or whatever, with people trying to use a technical product. So I think there's a lot of new opportunities as tech moves from just being the tech side of thing to tech- enabled sales and tech- enabled success and tech- enabled marketing and tech- enabled content. So pretty cool.

Avi Freedman: Absolutely. Even product and design, which is," How can we make this easier for people?" The platonic ideal is every support ticket is a bug. In practice, we don't... with enough functionality sometimes you can't actually do that. People have disagreements. But I think maybe we could do better as an industry on SaaS side too. And again, Packet was sort of... Yes, it was SaaS principles, but towards crosstalk-

Zach Smith: Yeah, SaaS- ish.

Avi Freedman: ...helping people understand what is the map of a SaaS company and what are the skills and overlaps? I'll definitely think about that as we do recruiting.

Zach Smith: Oh, and if you're listening to this and you're a new to career, just reach out to Avi or me on LinkedIn and bug us, because we will probably reply because that's something we do and don't feel shame to do that anywhere. Reach up, hold your hand and say," I want to learn more," and you'll probably get a fairly positive response.

Avi Freedman: Happy to do an intro. It might not be exactly our field. Well, we'll do it at the end also, but how should people reach you if they can't, if they're listening to this and need to reach you right now?

Zach Smith: The snail mail at... No, I'm just kidding. I'm on Twitter, zsmithnyc. Or, I'm on LinkedIn, which is zsmith. I managed to get that one. Both of those are the best. You can also find me in lower Manhattan, generally around, I drink coffee. So I'm usually at a coffee shop or something.

Avi Freedman: Quickly, crosstalk is not in lower Manhattan nor is hill country barbecue.

Zach Smith: Exactly. But those are the two best ways. Twitter, DM me or a mention, and LinkedIn is a great way as well and always happy to connect or learn more. I also sit on, like I said, the operating board of Pursuit, and participate in that pretty heavily. So it's another great place if you're interested in launching a career in tech or whatnot, transitioning from a different career path, that's a really a good place where I'd be happy to meet you.

Avi Freedman: Okay, cool. And I'm avifreedman at Twitter and LinkedIn and @ avi @ kentik. com. Hopefully the spam bots are not doing audio too. Oh, oops. I guess actually we do transcripts with this podcast. So that's one of the reasons I liked Casted when I was... Some podcasts don't do the transcripts, which I thought was like," Well, we want Google to find this stuff. In Casted actually you can highlight it and it becomes a clip and you can embed it in a blog and stuff like that. So it was pretty cool. So network and cloud, and as you said, there's all this transition going on. What's the difference between cloud networking, SP networking, enterprise networking? What you were doing at Voxel? What you're doing now? People think with cloud, there is no network, it's just APIs crosstalk-

Zach Smith: Right. Except that's the two things you get. The one thing you actually have to buy from clouds is the network. Everything else is pretty optional.

Avi Freedman: IBM, we have a friend there who says... he has two sayings. One, there's no cloud without network. And the more zen one is, network is the water of the clouds. It is crosstalk-

Zach Smith: I'm so one with you, Avi. Well, we think a lot about networking in a broad sense at EquineX. I mean, it's the core of our business, although six plus billion of our revenue comes from selling colocation and related. The reason why people do that is because they can interconnect at EquineX. And so networking is, whether it's at layer zero or above is where we generally derive our value and then a mission around creating connected ecosystems. I think that like most parts of the technology stack, networking is undergoing a really really seismic shift in most enterprises from effectively a pretty static, I'm going to call it back of the house operation, where it was in service to some pretty slow changing product cycles and slow changing business cycles. Kind of akin to a telecom product cycle in a regulated market. You don't reinvent MPLS every year kind of thing. And so there's a lot of networking that is built into a concept of a very long cycle and slow change, which allows you to have things like functional handoffs between the architectures and the operations. Those things can really work because you can create the runbook and the training manual and then train everybody on how to do that, and then that can subsist for a while. We used to have that in a software too, waterfally, creates software, ship it once every three years, install it and support it with the ops team who did the supporting of the software and the patching while other people wrote the software. And now we're in this continuous release cycle of whether it's SaaS or related where run it is the deal. Dev Ops, accessory model, I don't care what you want to know. But it's moving too fast and it's too complex to possibly do operations separate from the creation.

Avi Freedman: Right.

Zach Smith: And so I see that happening. It's like an echo right now that I see happening in the network side where there's a big transition towards a world network is pretty much software. Obviously there's a significant amount of other things that are required just like in software, you need the computer to run on it. But there is this wide concept of the transporting thing and whatever, but Network in terms of a capability set is software. And now that's starting to move to software speed, which is frankly, a challenge for some of the organizing structures of a lot of enterprises and service providers. Service providers got their faster because they were in the business of doing more of it. But enterprises now who are finding that digital is their core competitive weapon or their core competitive defense, whichever one you want to do it... Digital is the thing while networking is just one of those software tools. And so I think that's causing it to move faster and be part of the innovation cycle versus this back office. So my hotels or restaurants analogy is now it's moved front of the house. It's experience, network is the experience.

Avi Freedman: As we wrap here, any last thoughts, advice you would give earlier Zach? Would you still go to Juilliard? I guess that's the question.

Zach Smith: Oh yeah. I would still go to Juilliard.

Avi Freedman: Okay.

Zach Smith: That was great. It was my best choice ever. Got to New York city, met people who were world class at something, and tried to frankly learn how to be self critical. I would say ask more questions, it's just always helpful. And so young Zach, although asked questions, still had quite the... we all have our little egos and whatnot, and I would chip mine down. And the other one is I would probably of course, and this is probably what every 40 something year old says to their 20-something year old self is," Take some time, enjoy what you..." I was so jackrabbit, ready to get going. And I think in a lot of ways, some things both I did in life and business I passed by too quick. Well, Avi, thanks for having me. It's been super fun catching up with you and talking. And I'm glad that we didn't get too far into networks because you know, well, shaky ground for me.

Avi Freedman: We can do that next time. So again, Zach, how-

Zach Smith: So it's network as funk, right?

Avi Freedman: Network as fun, network as funk, all those things.

Zach Smith: I got you. Okay. Cool.

Avi Freedman: API for network as a service. How should people find you, again?

Zach Smith: Yep, zsmith on LinkedIn or zsmithnyc on Twitter. And I'm not going to give you my corporate email because it will most definitely get stuck in this spam.

Avi Freedman: Okay. Well, we won't do that. Well, thanks again. And thanks for catching up. And I definitely got some follow topics that we can nerve about offline as well.

Zach Smith: Thanks, Avi. Take care.