Tyler Jewell, Dell Capital

Thank you for tuning in to today’s full episode of the Breaking Changes podcast. I’m your host and Chief Evangelist for Postman, Kin Lane. With Breaking Changes we explore specific topics from the world of APIs, but through the lens of business and engineering leadership. Joining me today we have Tyler Jewell from Dell Capital. Tyler shared with me his unique view of investing across the developer tool chain, revealing how he’s been tracking the space for over a decade, and what some of the most compelling areas of investment are today and moving forward. Well, I always like starting with the basics. Who are you and what do you do?

Hey Kin, glad to be here. My name is Tyler Jewell. I’m one of the investment managers at Dell Technologies Capital. DTC is the VC arm of Dell Technologies. We focus on investing in disruptive technologies across a range of enterprise technologies, from silicon, network, storage, DevOps, data, governance, edge, 5G, IoT, machine learning, cybersecurity. And we’re one of the most prolific enterprise investors out there. We’ll invest about 300 million a year in those companies, and we’ve got about 140 companies in our portfolio, over 60 exits, and nine IPOs, which puts us in a pretty good range of VC performance. And we’ve got that performance because we invest financially around themes that we develop rather than on behalf of Dell’s businesses, so we’re independent, and the results reflect that. And I tend to specialize in developer and DevOps investments, which I’ve been doing for almost 17 years as an angel and at DTC. Prior to that, or at least during some of that period, I was a product operator for 25 years in and around developer tools and developer platforms at some really interesting companies like BEA, Quest, Red Hat, and I’ve also had the fortune to run three companies, two of which have been acquired, and one’s pretty large and profitable at this point.

Yeah, we met a while back at one of your startups, and then kind of stayed in tune over the years about what’s going on in the space, what’s happening, what’s interesting. And you always kind of captured me as someone really with your finger on the pulse of dev tools, and not just from someone who’s had your hands in building them, you understand what’s happening. You have a pretty sophisticated approach to tracking and understanding the space, and not just this year’s worth of trends, this goes quite a ways back. Can you talk a little bit about how you keep track of the space and understand what’s happening?

Yeah. When I was starting to do angel investments and investments on behalf of Quest, I knew I wanted to do them in roles and platforms, and my approach to figuring it out was to develop a thesis, and in order to develop a thesis I needed to see what all the current trends were. So I started building a database of every company that I thought was doing something commercially related to developer tools or platforms. It was private originally, and then I would just track how many people were working at those companies, what their revenue targets were, what I thought their revenues were going to be, and then how did that change from year to year. I started that in 2009, so it’s been 13 years that I’ve been building and maintaining that database, and as I come across new companies, either through introductions or discovery, I just always added it to this database and kept the data up to date, and then eventually published it for the first time, I think it was 2020, a couple years ago. The entire database is available online, but with my blogs that I do, in addition to saying, hey, here’s the database, there’s all sorts of interesting analytics that I can do in and around that information, and it tells me trends about which markets are expanding, which markets are contracting, where were the most new company introductions. There’s all sorts of things that you can glean from that, and it’s helped inform my own personal thesis around how to develop in this space, but also made a lot of great connections and discover all sorts of insights that I wouldn’t have had otherwise.

So you, I’m in the middle of my Gartner season as I would call it right now, and I’m doing the Forrester game as well, so I’m having to, working at a developer tool company, having to speak to these worlds and convince them of our strong bottom-up motion. And it feels like you’ve had your finger on the pulse of developer tools being relevant and important long before everyone else kind of understood or sees this. DevOps is a well-trodden road now, there’s a lot of investments and conversations. Why did you see the potential in developer tools so early on, when everything else was a top-down vision being pushed by Gartners and Forresters and others?

You know, I don’t know if I saw the opportunity or the vision. To be honest, I started doing the work just because I love developer tools and platforms, and earlier in my career I had built some, and I had tried to get the companies I worked for to do more with them. I had repeatedly run into roadblocks with that, and the most common refrain was, hey, there’s no money in developer tools, and that seemed pretty frustrating to me, because here you had a talent class, a very large talent class and group of people who were creative and innovative, and whenever they saw problems they would most likely just roll their own, they would want to go build software that solved that problem. So it seemed to me that classic business people who were dismissive of developers as a talent class were misunderstanding the opportunity there. A lot of why I put all this data together is just to help educate others on, look, there’s obviously more going on here than what meets the eye. And in many ways I’ve been fortuitous, because now the market treats it as a top-tier investment class, and they caught up to me, and I guess I’ve been the beneficiary of that. Most of my investments over the past 15 years have really done well as long as they’ve kind of aligned to this general trend.

How do you feel you stay immune to the trendy aspects of any investment, but primarily, I feel like with dev tools we’re starting to see more hyped-up trends that may or may not be real. How do you stay immune from short-term trends?

Yeah. There’s a lot of investors who are chasing what I consider historical metrics, like, oh, look at the GitHub stars or the growth in the GitHub stars and whatnot. I think those are superficial metrics, and they’re also backwards-looking, and they’re not necessarily reflective of whether there’s a real opportunity there or not, so you’ve got to be really careful with that. What I do is, first, I have kind of a 120-year view of the software industry. I know this sounds pretty wild and it’s pretty crazy, but in terms of software construction, our industry, I look at it in terms of three waves, each wave being 40 years. That first wave, which I call the waterfall wave, ended really at the year 2000, and we’re in the second wave now, and I call this the agile wave, which is the wave of making changes quicker and getting feedback on a more continual basis. And I have a hypothesis that we’re going to have a third wave that starts to kick in probably around 2035 or 2040. I call it the third wave of industrialization of the software supply chain, and that’s going to be a data-driven approach to constructing and maintaining software. Each of these waves are really 40 years, and I think there’s plenty of evidence that why these things take 40-year cycles. That’s my big picture. Good investors, I think, have to think five to 15 years out, and so right now I’m looking at, hey, we’re on the peak of the second wave, the peak of the agile wave, so I think we’ve hit peak DevOps, and I have a lot of hypotheses and theses about what’s going to happen over the next 20 years as this wave starts to wane and we start to see the emergence of the following ones, and I try to invest around those themes.

So is there new investments that you can make to keep writing and investing in the agile one, or are most of the investments that’s going to happen for the next decade going to be fueling into the data phase, or is there an overlap, do they work together, as far as the agile’s going to produce data and what we need for that next wave?

Yeah, I mean, I think when you think about these 40-year waves, they’re broken into two epochs. The first epoch, the first 20 years, is a fragmentation epoch, and that’s actually really happened here. When I started doing this database 13 years ago there was only a couple hundred companies in the entire database, now there’s over 1,300 companies, and these are ones that have not gone out of business. That database has doubled in just a few years, so the total number of companies certainly reflects just the overall interest, but it’s also reflected that generally we’re seeing fragmentation of the platforms in and around the agile concept. The second half of the epoch is a consolidation epoch, and so in that sense there should be an overall reduction in the number of companies, favoritism given towards platforms or systems that really unify a bunch of complex technologies, things that focus on automation rather than piece parts. And that’s investable, and that consolidation wave will take 20 years to play itself out, and there’s lots of pockets of opportunity for really big companies to emerge that really play to those capability sets. I do that, and at the same time I’ve got my eye on what I think that third wave will be, that starts in 15 years. Something that starts in 2035 or 2040, that’s a long ways off, that’s a bit risky to put money towards, you need to have a 20-year investment horizon, but you never know, and if the right team and product and technology come along that plays to that, I put money to work and try to help them get there sooner.

Well, on the data aspect of that, my background’s database administrator, so ‘87, ‘88 my first database work, and so I’ve ridden the waves of data, and the different ways that happened, I felt like we were pretty played out data-wise until recently. I started seeing data at a scale, and it’s not the Hadoop and kind of these data tools, it’s the data exhaust from some of the operational-level stuff you talked about that are contributing to agility, performance, velocity. So it’s CI/CD pipeline data, how do you help developers be more, optimize there. It’s gateways, as the gateways are commoditized, how do you measure velocity and release there and optimize that? GitHub Copilot is not about the AI or the ML, it’s about the data and what you learn there. So what are you seeing on the data ops, ML side, what’s interesting with data right now that you feel is kind of priming the pump for this future that you speak of?

Well, you created that, the data being incomprehensible at the point, I think that level of scale has caused the data industry to rethink what are good data structures in order to manage all that sort of data, that’s one driver. The second driver is that there’s a theory out there that says, if every company is going to be a software company, then the only differentiation you have as a company is going to be the quality of your software, and most companies are going to be able to produce software of an equivalent quality, so how do you differentiate? You tend to differentiate through your machine learning algorithms, and with your machine learning algorithms, guess what, the models are all probably going to be commoditized, but the data set that you used to train those models with might be proprietary and unique and ultimately make one product better than the other. So you look at that and go, hey, if it’s machine learning that’s going to be the fundamental differentiator of all software in the future, then my personal proprietary data set that I have, which is feeding into that, is my entire company, and so I need to hoard the data, I need to generate a lot more higher-quality data, I need to make sure it’s cleansed and that I’m using the right models which are completely explainable. So that’s a huge catalyst there. And so to your specific questions, I think the amount of data and the different types is driving a rethink of data architectures, a whole new category of data architectures, and we’re in the earliest stages of that, there’s tons of VC money that’s flowing in there. And then because machine learning is going to drive so much, I think ML ops is certainly a category that organizations are struggling with, because there’s no winner there, there’s dozens if not, there’s like 35 to 40 companies that are trying to play around in the ML ops space, whether they’re focusing on developers or maybe the data scientists or whoever it is, but there’s no winners, and they’re all trying to figure out what is, if you will, the agile for ML ops, what’s going to be that standard workflow, and is there going to be a platform for this. And right now it’s an all-out battle to see who’s going to establish that at the end of the day.

Yeah, and like you said, a lot of it’s going to be the model-based, so ML ops but also model ops, how do you optimize the usage of these models and evolution of them, but with an eye towards your data. And I know a lot of folks, we had Ford on last season and they talked about the data opportunity for training models on fleet data and just driving around on a daily basis doing deliveries, trying to make sense of that world, and equipping folks with the ML skills they need to do that, to generate the data and iterate on the models. So do we have the skills we need in the enterprise to be doing all this, because it seems like data scientists are a pretty hard group to cultivate, you can only produce so many data scientists. But do we have the ML skills and the data skills we need to get to the next level with this?

I think the whole industry is catching up. ML and ML ops as a skill set, since we don’t have standards on what the right processes are, if not productizing it, they’re definitely experimenting with it, and as we’ve seen from other waves of paradigms that get introduced, it’s a good 10 to 20 years before organizations mature in how they apply a new paradigm to it. So I think we’re in the earliest stages, for what it’s worth, we’ve got a long ways to go, and that’s pretty darn exciting. There’s certainly not as many data scientists and data engineers on the planet as there are software developers. Now if you are doing data science for the purposes of analytics and you can work in a silo, that’s great, but if you’re doing data science and machine learning because you want to have inference capabilities or other sort of capabilities that are features of an application that your developers are working with, you’ve now just compounded the problem, because now you have to intermingle what is an agile release process for software coupled with a machine learning model that’s continuously going through training of data, that also is getting monitored, and whose tuning could absolutely change the behavior of the application itself. So you now have these two teams with these two completely different processes that have to figure out how to work together for the common interests of your end users.

And it feels like the tools are what’s going to matter here, because I’m not going to be able to get my team developing all the skills in all these areas they need, we need the tools to augment them with the abilities they need.

I think it’s a little bit of the tools for sure, tools that help create guardrails and facilitate process collaboration. I also think that there’s a lot of knowledge of how you manage your data, which has nothing to do with the tools, the quality of the data, the freshness of it, how often you’re dealing with training. That’s just going to be something, guess what, developers and data scientists have to become specialists in this now whether they like it or not. And then there’s an infrastructure problem too, which is why Databricks is so valuable, because just having the data and having the model isn’t sufficient, training that model and doing it at scale, there is a responsiveness of that and then the cost of the scale-out infrastructure needed to do that. And frankly you have to be at a pretty significant scale for these models to get accurate enough where people feel comfortable about putting them out in front of end users.

And this is, in our lack of a better word, our grandfather’s AI, this isn’t Watson, massive brain, you just feed a bunch of data into it. This is much more modular, flexible, ever-changing, this isn’t a destination, oh, we have the model trained, we’re good. This is a perpetual motion going forward.

Yeah, so models age, and then the data that you used to train that model with.

So is there life cycle and governance concerns? I’m neck deep in that with just APIs, but is there a whole life cycle and governance way of thinking that’s emerging?

Yeah, there’s certainly a bunch of vendors who are just focused on the explainability of AI. If you’ve got an inference and it gives you a result, what was the quality of the data that was used to infer that result, was the data legal, was there no PII information in there, how does this result compare to the quality results given to other people. In some cases you have to have traceability of how did the algorithm conclude that this was the result. There’s any number of issues that are on that. And you also then need to be able to trace quantifiably how an algorithm is degrading over time, because it falls off an asymptotic sort of curve there. So you’ve got to monitor that, and if it falls below certain thresholds then you’ve got to make changes or remove it or whatever that may be.

Yeah, you touched on observability, it feels like there’s going to be, you mentioned traceability, it seems like observability at this layer is going to be pretty key. But are you seeing regulatory creep into this as well, is there anything that’s government compliance that you’re seeing going to influence how we move forward?

I’m sure there is, but I haven’t been monitoring that.

Yeah, I’d like to tune into that, because I feel like data-wise it’s definitely a concern. API-wise we’re starting to see banking, healthcare regulations emerge as far as what you can have. But it seems like with some of the backlash around ML that there’s going to have to be more observability and compliance. Anytime there’s a piece, if they have data there and they’ve influenced a life decision, it’s inevitable that it’ll have to be regulated.

Yeah.

Now I have to think about your 40-year model and maybe spend, you know, 1960 through 2000, and go, what tech regulation existed, and then bring some of that research, share with your database, see what I can contribute there. That’s something I’m trying to tune in more and understand. So we talked a little bit about the skills and the resources and the tools are going to be augmenting people. Do you see opportunity in here for non-developers, because there’s only so many developers in the world, and it feels like we’re going to need business stakeholders, stewards of some of this data from a business domain being involved in some of this work. Is low-code, no-code getting the traction that you think it should, that some investors are saying that it has?

Well, one of the premises of what happens between a fragmentation epoch and a consolidation epoch is that while the original founding principles that drove that wave have now been validated, through the fragmentation process you’ve now introduced unnecessary complexity around a bunch of different tools and processes that optimize individual components of that principle. So this idea of peak DevOps also kind of implies that we’re at peak complexity for what it takes to build, construct, maintain these software systems. And even though a developer today can build something in an hour that would take them many days to have built two decades ago, the rest of the process, the communication, the tools, the languages, everything else, and maintaining that, may have introduced a lot more complexity than they were willing to bargain for. So how do you address complexity? Complexity is addressed through automation, and there are over a dozen different forms of automation that exist in and around the DevOps world, and one of them, probably the most dominant one, is low-code, no-code. It’s wild, I think there’s something like 130 companies in that database of 1,300, 130 of them we’ve classified as low-code, no-code companies, and there’s more than a handful of them that are well over 100 million dollars of ARR. And there’s a pretty decent set that’s between five to twenty million dollars. So you’ve got a lot of companies, instead of one or two dominant ones, you’ve got a whole pile of companies who are really doing well at scale, and all the indications are that these companies are just getting warmed up, that there’s a whole category or class of systems that need to be built that people would just prefer to do with a low-code system as opposed to a traditional way.

You feel like the notion of what is an application is going to kind of be blown away in this new reality, because historically, I usually paint the picture as web apps, you have desktop apps as well, then we had mobile apps, and then now we have some more device internet of things, but then the network is programmable now. So do you feel like the notion of what is an application is radically different now?

Oh, you know, not personally, no, but I’d hazard your listeners probably do feel that way. What I do see is that I’ve always said that APIs are the only universal application that exists out there, it’s the only universal GUI. And most systems that are going to be built are going to first be built as an API, and then you might have some clever user interface mechanisms to go along with that separately and independently of that. And I did some back-of-the-hand math, and my guess is if you take everything that’s getting exposed over the internet that is also programmable, however you define programmability in this regard, by the end of this decade there will be well over a trillion live active programmable endpoints on the internet. Whether it’s your phone, or some hosted API provided by a vendor, or it’s an IoT device that has some accessible endpoint, you pick your poison filling in programmable endpoints, that’s a lot of services that are going to be out there that have to be managed, and frankly low-code is probably the best way to connect to and integrate all those different points.

Yeah, platform scale’s the only way we’re going to discover those things, know where they are to even use them, how we’re going to put them to use, how we’re going to monitor their health and reliability, observability, traceability, know when it breaks, all of those things. Because the number of APIs I use is just through the roof, it’s not hundreds anymore, it’s not tens or hundreds, it’s thousands of APIs, and I cognitively can’t keep up with that, so I need tools to help me make sense of this world.

And on that point, you can’t cognitively keep up with just the number of APIs that you’re using. How are developers post-cognitively going to keep up with a modern-day system when that modern-day system is microservices, which are lots of different independently moving parts that are all potentially communicating with one another, a heavily fragmented tool chain, an agile process that encourages you to make more changes because you can get more feedback, so now you have to do continuous feedback, which is a lot of data to observe, comprehend, and process. And because there’s now 1,300 different vendors out there, you have to have some reasonable expectation of studying what’s new and keeping up and educating yourself. So one of the things that I look at and go, how can a reasonable person comprehend the impact of a change that they want to make and then make changes on a continual basis in this world? And I think that it’s ultimately going to lead to a path of not being sustainable unless you have a really small system where you can keep it all straight in your head, but any sort of system of reasonable complexity, it’s going to be difficult to maintain some organizational structure. And you go, well, that just doesn’t strike me as a good thing for our industry. So with rethinking our fundamental principles and having machines do the change reasoning for us, with intent provided by humans, would that be a better overall approach on how we build and maintain software systems versus the way we do it now?

Yeah, that really touches on a whole lot of the world that I see moving and shaking right now. Because back to the platform argument, I can’t make sense of this world, I can’t understand all the direct services being used, let alone the dependencies of those services. One of the conversations I had yesterday was around breaking changes, the title of this show, but how do you know if I add a property or change this property, what are the impacts, who’s going to scream, who do I need to talk to, how do I do that preemptively in a large-scale system? I can’t do that today in most enterprise environments.

Now in the data world there’s data catalogs, which are basically large systems that aggregate the metadata about how those data systems are used and how the data is consumed, and it’s amazing to me that we don’t have the parallel in the developer world, which I call a service catalog. I’ve written about this, and it strikes me as inevitable that much the way that Alation and Collibra have now become these billion-dollar unicorns that focus on data governance, that we’re going to need to have service governance really emerge. And in service governance, the product would be a service catalog, which is the aggregation of metadata not just about the process and the teams, like Backstage from Spotify, but metadata about the systems themselves, their communication paths and where they’re deployed and how it’s architected and all sorts of stuff like that. And that metadata can inform people to make better decisions, but it can also inform the tools as well, so that the tools can start taking on some of these decisions and understand the potential impact as well.

Yeah, I see a couple startups emerging, I’ve been a part of several of these, whether it’s discovery, I don’t like to use that phrase because a lot of people think search engine, but discovery from a circuit-breaker pattern, service mesh type, automated understanding of not just the interfaces and the services, but as you said, the operations around them, the teams around them, the reliability, the overall scalability and health, and whether I should be depending on this. Because if I’m going to introduce it into my system as a dependency, I need to know, I need some guarantee, some assurances that it’s reliable, not just in SLA, I need some track record, some history that I can go on, so it’s pretty important.

Exactly.

Yeah. So I’ve seen several folks come along with ratings-agency folks, for lack of a better phrase, Standard and Poor’s but for services, trying to understand the wealth and the health of this catalog and what’s possible, but I have yet to see something emerge that helps us properly manage risk at this level and understand it within just the services we depend on, let alone understand the myriad of third-party services and other external dependencies that we have. So what gets you excited about investing today? Aside from the investments you’ve made, what are you getting worked up about and thinking is going to be hot next?

You know, I can do a little bit to try to make those things happen, find those companies. But the big thing that gets you excited about investing today is the teams that you invest in, the people that you get to work with. Early on when I started investing I just invested in the idea, and as I’ve gotten older and you realize that time is really precious, a single investor really can only do deep investments in 10, 11, or 12 companies, and you’re going to spend a lot of time, hundreds if not thousands of hours, with the founders, helping them achieve their dreams and their desires and their vision. So you really gotta get excited by the people, believe in them, and see yourself as part of their team, and that’s what you look for.

Yeah, agreed. The ideas don’t matter much to me, I’m always looking for people who don’t just get the tech but have a passion for whatever they’re building, have a genuine curiosity, interest, a certain amount of work ethic. I would say COVID has kind of revealed for me a lot of what our motivations and incentives are all about, and some people who may or may not be in it for the long run, we’ll just say, and others are struggling with delivering at a scale I think that’s needed today. So for me it really comes down to the people. I’ve seen enough APIs, I’ve seen enough startups, successful, not successful, it really comes down to the personalities and the people and the stories that they tell about what they’re building, how they’re building it, and what’s going to matter.

So most people who take the risk and the leap of starting a company almost universally come across as great, work out, very passionate, willing to climb mountains. I’m curious, do you come across founders from time to time who really got into it for the wrong reasons, and they’re not really there?

Yeah, I’ve seen a few that believed they had the right technical solution, and they learned the hard way that that doesn’t always exist, let alone apply at scale in whatever industry they’re doing, and they were unable to come to terms with that. So they thought they had the right technical answer and that it would just translate into business revenue and it would just make sense, and it didn’t, it needed a wider story, it needed much more than just the technical. But then I’ve seen some who are just doing it just because they want to be CEO and they don’t really care about the thing. But I’ve also seen a lot where they’re really believers and they buy in. That’s one of the reasons I’m at Postman, is because that exists here, really trying to understand what people need and trying to find, we just celebrated yesterday a decade of when Abhinav, our CEO, posted the Chrome extension on the Stack Overflow message. I tweeted it out yesterday, but it was that first Chrome, hey, we’re trying to solve a problem for developers here, you can’t see APIs. So it’s not API testing, it was troubleshooting, trying to make sense of this very abstract invisible world with this Chrome extension, by adding this layer to what you can see, and now it’s become this whole platform that does the same thing but at scale. How do you see APIs, and I think anybody who really cares about helping folks see this digital, any part of this digital transformation that we’re in, is going to be valuable, but it’s got to be someone who’s willing to do the mining, the hard work, to understand what it means, and out of all of this noise, what’s the actual signal, what matters the most. And those are the people I think that I would be investing in, but I’m not an investor. So how’s COVID changed anything for you, has it changed any of the way you work, the way you interact with people, what’s changed?

Well, the idea, pre-COVID, of outlaying millions of dollars into a company without having spent a lot of time with the people, not just on phone calls but at their offices, in social environments, getting to know the founders, that was just a non-starter. The idea of not being able to study the individual, build the relationship, and go through those normal mechanisms prior to putting that much money to work, it was incomprehensible. And through COVID now the entire life cycle has been done digitally, virtually, and we invested, gosh, probably 400 million dollars over the past 24 months, and didn’t need anybody, didn’t shake anybody’s hand. So that’s a pretty wild concept. I hope that it’s going back to normal. Our office just opened up last week for the first time, we were all there, it was wonderful to hug my coworkers, we actually had entrepreneurs come into our office and get to meet. The energy is just very different after two years of having to do everything digitally and remote. Putting people into a room, having a whiteboard available, there’s a tangible feeling that you can get, and so you definitely get a different read. So I hope we come to some sort of hybrid understanding of this, I don’t know what that looks like, but hopefully it’s not all Zoom and it’s not all in person, and there’s something in between.

Yeah, agreed. I did my first, the Stripe team was in town and they had a little shindig, I went to it, it was the first in-person thing I’ve done in a while, and it was interesting, it was good, but it was definitely going to take some practice I think for me to get back to it. So I’m fascinated by your database strategy, how you manage your information. How do you stay in tune with things, what do you use to gather information and stay aware of what’s going on?

Every morning I read through a list of hundreds of feeds from companies and smart people out there. That gives you lots of insights, lots of threads to go pull and chase, lots of discoverability that comes from just doing that exercise. Two, because of who I work with, I get updates on all the different fundraising events, and so that gives me signal to companies that maybe I wasn’t aware of before. Three, a lot of networking with other investors and other interesting entrepreneurs. We have over 100 entrepreneurs in our network, founders, and they also are looking at the ecosystem and giving us insights. And then I’ve also got 15 partners here at Dell Tech Capital who are doing similar sorts of things. So all that together leads to a bunch of data points that are coming in each and every day.

I like it. Yeah, I’m not good on Twitter, there’s a lot of people who are fantastic on Twitter, but I just find that overwhelming, but I think there’s some investors who just live by that, that’s how they build all the relationships.

Yeah.

I’m good at the Twitter, but I have to say there’s some diminishing returns there. LinkedIn, I’ve been pushing on them, and they’re doing a good job of evolving their API to make it as robust as Twitter’s so that I can understand the signal versus noise on LinkedIn, but LinkedIn’s kind of trumping me for tapping into what’s happening. It’s been interesting, I didn’t anticipate that after the Microsoft acquisition. Well, I think we covered most of it that I wanted to. I really appreciate your time today, this was enlightening, I love hearing your view. I’d love to get together in person again at some point, be able to keep talking through the space, because I’ve got some other startup scenarios I would love to explore with you, but I really appreciate your time today, thanks for sharing your approach and how you see things.

I’m glad to be here, it’s great to hang out with you, Kin.

Yeah, anytime, I’ll be in touch and see if we can’t bring you back for some other conversations, but if you’re seeing anything interesting feel free to ping my way as well, and until then, I appreciate it.

All right, sounds good, take care.

Thanks again to Tyler for stopping by. You can find more about Dell Capital at delltechnologiescapital.com, and you can find Tyler on LinkedIn. You can also subscribe to the Breaking Changes podcast at postman.com/events/breaking-changes. I’m your host, Kin Lane, and until next time, cheers.