Vatsal Shah of Litmus Explains What It Takes to Create High Impact Solutions
Transcript of Vatsal Shah of Litmus’s interview on the Manufacturing Hub Podcast
As part of the 4.0 Solutions community, I wanted to do my part in helping raise awareness of the whole Industry 4.0. So, I created a word-for-word transcript of this fascinating interview with Vatsal Shah of Litmus on the Manufacturing Hub Podcast.
It took a while to eliminate all the errors in the automated transcript. But it is my hope that others interested in learning about Vatsal’s mission can now find and consume this content more easily.
If you are keen to learn more what Vatsal is doing, plus other Industry 4.0 topics in general, I recommend you join visit the Litmus website, where you can get all your questions answered!
The source video is here:
Transcript
Dave Griffith: Welcome to the Manufacturing Hub podcast. We are a little bit earlier than normal. So, as you guys are in the chat, please let us know if you like this noon East Coast time. Waking up with your coffee West Coast time. Unless you’re Vatsal, who is like always 120% awake no matter what time of day I talk to him. Yes, you guys must have some special coffee out there in the Bay Area, Vatsal. But no, let us know your thoughts on the noontime if you guys like this. And if you guys are new here, welcome. If you’ve been here before, welcome back. If you’re new here, on Manufacturing Hub, we like to have kind of an open-ended conversation. Every month we pick a topic. This month we’re talking about delivering high impact solutions quickly, sponsored by Litmus. You can see the logo up top and on Vatsal’s shirt, both the same. We’re going to get a little bit of behind the scenes, especially what it took to help build this platform. And perhaps, for folks who don’t know Litmus, what exactly Litmus is and the problems that they solve. Fun fact, they solve basically every problem, but we will absolutely go ahead and get into that in a moment. And then, before we go ahead and officially jump in, I do want to let everyone know that I will be with, I guess with Vatsal, and the Litmus team at Hannover Messe. If you guys are there, we will be in Hall 17, Booth F18. And I have learned that it’s exceptionally important to say all of those letters and numbers because otherwise, you’ll literally never find it. No one will ever be found. So absolutely, we will go ahead and talk about that. But please feel free to reach out to me and Litmus. Come hang out with us in the booth, anything along those lines. Without further ado, let’s go ahead and kick off. So everyone, welcome to Manufacturing Hub with me, Dave, and this guy up here, Vlad. We have a very special episode this week. And this month, we’re talking all about delivering high impact solutions. This week, we’ve got Vatsal Shah from Litmus on the show. Vatsal, thank you for being here.
Vatsal Shah: Oh, thank you for having me. Thank you, Dave. Thank you, Vlad. It’s great to be here. And I listen to you guys every week and very glad to be there. Thank you
Vlad Romanov: Really appreciate that. Thank you so much for taking the time to talk to us this morning. But, so before we dive into the technical topics, I wanted to get a little bit more of a background. How did you get started in your career? What was your journey like in manufacturing automation? And ultimately, what brought you into starting Litmus?
Vatsal Shah: Okay, that’s a long question and a long answer, but I’ll get started. So, my career started as an industrial automation engineer. I was the one who was designing PLC logic, ladder logic, and SCADA screens. I was installing, doing F80 and UATs on the plant. That was my first job out of my bachelor’s, my university days. I did work on various food and beverage, a couple of oil and gas, and other customer projects. Essentially, I used to read the PID diagrams, try to replicate the logic on the PLC side, and try to copy-paste the SCADA screens from previous references. But that was the starting year. While working on that, one thing that I personally saw is there was a growing need for a heterogeneous environment. There was a growing need for data coming out of those heterogeneous environments. For a team of four engineers, it took us more than a three months’ time frame to integrate Rockwell, Siemens, and some of the Yokogawa system together and push data into a single SQL Server database. So again, of course, we used all the fancy tools in between and wrote some Excel VBS scripts as well to transform the data. But, starting from that, the idea was we wanted to create one platform which can talk to every control system or every industrial system, be it hardware system or software system, out there. We aimed to bring data to much smarter systems in a matter of seconds or minutes. We were ambitious to begin with, we didn’t know what we were up against, but that’s how the company and that’s how day zero got started. A lot of smart people joined over a long period of time from a manufacturing industrial background. Where we are now is really that common data layer which is powered by industrial edge computing.
Vlad Romanov: If I can ask you, Vatsal, I’m really curious about the early days of Litmus. I think that as an engineer, we sometimes underestimate what it takes to bring a product to market. So, I’m curious, what was your initial MVP like, if you want to call it that? That customers saw value in and sort of wanted to try it because I think the goal of getting all the data is extremely ambitious. We all understand the complexity of the production floor, but I’m curious, what did you start with very early on just to demonstrate a proof of concept?
Vatsal Shah: Very early on. So, I think I jump into MVP, but let’s say in the early days, John, who is my co-founder, and I, we were trying to reach out to early customers. We were just trying to ask them if this is the idea that you would ever purchase. And this is the idea that we got. I would say that 90% of them told us, “This is never going to work.” Those were the first six months, which was continuous rejection. They said, “You guys are too small, you don’t know what you’re going to do. You’re going against historians, you’re going against this company and that company.” So, it was always there, but we kept on understanding their requirements. We kept on asking them why it will not work, and we tried to build or craft the product and stories around it. It’s always a chicken and egg problem. Unless you have the first customer, which is large enough or a marquee customer, you’re never going to get your second customer. So, MVP was that. MVP was just crazy. It’s like nobody will ever buy it, and I don’t blame them. That MVP was the idea that we are going to build all the drivers. Sure, so that was my credit card, and I purchased a control logic system at day zero. That was initially what we did and then tried to create a protocol driver which was streaming data to our central cloud platform directly. So, our MVP was not Edge. Our MVP was a cloud platform and we were giving our Edge for zero dollars. Seriously, our Edge platform was no cost. It’s like we want to bring all the data to the cloud environment. That was MVP, and in two years in the making, we shifted our cloud-first strategy to an edge-first strategy. There was a big pivot for the company because customers started asking, “Oh, your free product is all I need. I don’t need your cloud product. I just need all the drivers that you have, Vlad, everything that you have in your back wall.” Like, for example, we have the same in our Santa Clara office. It’s like a wall of 1970s up to this point. Every control system that we can purchase, we’ve already purchased it. So, customers wanted those drivers and protocols for free, and that made us change the business model. Which is okay, if everybody wants the edge-first approach, we might as well create an Edge product rather than a Cloud product. And that is why we are here. So, where we started MVP, that was more like a learning exercise. Where we are now is an amazing product-market fit.
Vlad Romanov: Why was there such a hesitancy on the cloud? And do you see them being a little less hesitant on the cloud side in general in manufacturing nowadays than when you were getting started?
Vatsal Shah: A huge shift happened. Around the 2015–16 timeframe, the cloud was still early. The technologies, the privacy policies that they had, and the security controls that they had, they were very naive, or they were at infancy at that point. All the cloud vendors, or hyperscalers, they invested an aggressive amount of money over the last six to seven years to build trust with their customers and partners. And now, the customers are less hesitant about going to the cloud than they were in 2016. But the technologies have matured, their processes, their business models, security layout, and the general architectural posture for sure have improved a lot. So yes, there was a lot of hesitation before. Giving your data to a small company, pushing data to the cloud, was not a smart idea to begin with.
Vlad Romanov: Dave, any thoughts?
Dave Griffith: Yeah, think that’s very interesting. Especially in the 2013 to 2017 range, I also experienced cloud hesitancy, as we can call it if we want to be nice. Most of the time it’s, “Heck no, we’re never going to leverage the cloud on the manufacturing floor,” because it’s something different than what we have done in the past. I am happy to see more of that adoption now. I guess, Vatsal, my one question for you is, you’ve kind of walked us through the beginnings, through MVP, through the edge of Litmus. Today, if someone is looking for Litmus, or if we’ve got listeners who are like, “Hey, I don’t know what Litmus is or what Litmus Edge offers,” what is the best way to answer what you guys at Litmus are now, as your platform and the services that you guys offer?
Vatsal Shah: Sure, so Litmus is a company. We are an industrial edge computing software pack. Our software is deployed on the plant floor. Our software is installed on level two and a half to three of the Purdue model. We collect data from every industrial system out there. We develop native protocols and drivers for Rockwell, Siemens, Omron, Yokogawa; it doesn’t matter. We have the highest amount of coverage. Vendor number two is one of the legacy OPC UA servers. Them, and a few others combined, they still can’t match up with us. So, we have the highest amount of connectivity in the industrial system. Now, once we started collecting data, the next step was the normalization or contextualization of the data at the edge. So, in this simple web user interface, as soon as we collect the data, we normalize it. We contextualize it. Then we added a workflow engine. After that, we added an analytical system. Then we added a machine learning or container system, and then we push data into the cloud or any other system that they are looking for. So, we are the software platform that enables our data-driven innovation journey for users. We collect, process, analyze, and integrate all of the plant flow data in an air gap, first offline cost environment. When they are ready, they go to the cloud.
Dave Griffith: Interesting. Okay, I love that. I feel like that is certainly a solution that is needed. Especially if we can offer it edge-first, air-gapped first. That solves many legacy struggles that I have seen of groups that are like, “Hey, we want to go leverage a piece of technology, but we are unwilling or unable to go connect it into the cloud or into the internet in general to leverage that.” So, I think that’s very interesting. And then, I guess I’d love to know, today, do you guys have a target customer? If we’ve got people listening saying, “Hey, that sounds really interesting,” are you guys talking to every market, every vertical? Are there certain sizes of companies that fit Litmus best? What does that look like for you today?
Vatsal Shah: So, today, we are deployed widely throughout the world. North America, EMEA, and Asia Pacific markets in general. We have a large number of installations. There are 2.8 million endpoints connected with our systems now. We are in the market directly, plus in partnership with specific OEMs that white label us. We are in the market with hyperscalers, and we are in the market with a system integrator network as well. We pretty much cover a broad spectrum of use cases. Where we shine, in ascending order, I would say like we have a good footprint in automotive, tier one, our automotive customers. The second one goes in food and beverage. The third one goes in high-tech electronics assembly. Recently, pulp and paper, chemical, biopharma, are more going towards process as well. We are really good on discrete or hybrid installations because they have a large number of heterogeneous assets throughout the plant flow. And when you have distributed assets, when you have a heterogeneous environment with different types of assets, you can install Litmus Edge in multiple networks, consolidate all the data, because we collect and normalize it, and bring it to a central environment. So, in discrete and hybrid environments, we shine right away. In the process industry, yes, we are learning. We are growing in the market with a partner portfolio.
Dave Griffith: Absolutely. So, (1) I love this and (2) I’m going to take Vlad’s next question of, “Oh my goodness, that sounds awesome. I would like to go spend six hours to get a demo,” and remind everyone that we will be at Hannover in a couple of weeks. If you guys are interested in some live demos, come on over to Hall 17, Booth F18CI. I remember it now because I’ve said it so many times. Come check out; there’s a bunch of live demos. If you guys aren’t going to be there, you will be building with Vatsal, and we will be doing some live builds. Assuming we’ve got internet connectivity, we will be streaming to you guys, our listeners, so that you guys can see as we go and build some of these solutions to get an idea of what that looks like. I will say that if we have technical questions beyond that, there’s a high probability we can get Vatsal and someone from his team on, in order to give us maybe a bit better of a demo at a later point in time. Having said that, we kind of promised people applications and solutions. We build this as how to deliver high-impact solutions. It certainly sounds like you guys have the platform to be able to do that. Do you have a good example or two that maybe we could talk through to hear how people can leverage the Litmus platform to go deliver those solutions and how they can go deliver those solutions quickly?
Vatsal Shah: Sure. So, time to value across different use cases is one of the biggest priorities that we have. It’s one of the fundamental principles that we have. What we tell our customers is that within the first hour, you are already connected to all of your assets on the plant floor. Within one day, you are already analyzing that data. Within one month, you are already deriving KPIs or ROI out of the platform. So, what we learned, as soon as we launch a platform or a feature or a product, we get in front of customers and we ask them how they are using the product. How they are exploring different use cases. We had continuous learning over the last three to four years, where we go back to them for asset performance monitoring use cases, some of the predictive maintenance use cases, recently machine learning-driven predictive maintenance. There are modern use cases which are driven by augmented reality, some cloud-connected MLOx workflows. More and more use cases are opening up. But to mention about solutions itself, there are vertical-specific solutions that we have seen that are successful in the market, and then there are horizontal solutions which are very successful in the market as well. There are both sides of the ecosystems that exist now. But how do we go about building it? We do start with a product-first approach. Our largest customer and our smallest customer are using the same version of our product. So, everybody gets the same foundational layout. Like customers, they get started with low investment, high impact type of use cases. What are those? They have a lot of different assets and they wanted to create specific alerts on certain scenarios. Those alerts might be coming out of those assets or PLC variables, DC drives, robotic systems. They start creating the workflow after understanding that data, and they create their visualizations. They create the alert workflow. They create the whole journey to cloud. Once they have it for a specific asset, they like to scale it up to multiple different sites, multiple different assets in general. So, those asset performance monitoring type of use cases, enabled at scale, are extremely low hanging fruit. Then, once they have enough data, they start going towards more complex use cases. They might be workflow or light soft manufacturing type of workflows or converting all their scheduled maintenance into predictive maintenance workflows. Those are extremely high impact, but it does require subject matter experts to spend time on top of the data because you can’t just build it out of the box for every production line out there. So, we bring all the data, we give them tools which are called “Ready analytics”, and then they bring those specific workflows. Then we go one step further which is energy monitoring, energy usage reduction type of use cases, quality assurance use cases, wastage reduction use cases, inventory and supply chain use cases. So, once you have access to the data, almost everything is bolted on top of it. That’s how we have built repeatability inside our platform. And of course, there are vertical specific solutions, but these are more horizontal solutions which are repeatable across customers. It’s a journey.
Vlad Romanov: Vatsal, I’m interested in a number of things, but I guess let me ask you sort of like a first question on the architecture side. So, you’ve mentioned that you’re a software-first platform, which means to me, at least, you can deploy that on any type of, I’m assuming, Linux machine. So it’s going to be either like a rack-mounted server at the facility, or can it go as low as, let’s say, a Raspberry Pi or like an IPC? What does that look like? And then, ultimately, my assumption, based on what you’ve described, is you typically set it up on, let’s say, the IT side. It still needs access for the OT infrastructure, and then it has a connection or a funnel still to the cloud where you can do some processing, some visualization as well. Like, what does the infrastructure that you set up typically look like?
Vatsal Shah: Sure. So, one thing that we realized early on is every network and every plant flow is different. It’s not a Greenfield environment. Sometimes they have a network which they purchase from Best Buy. They purchase and they install it on the plant floor. We actually see the same IP address five times on a plant floor which is like 192.16 at 1.00 is my Rockwell PLC. It has been a common practice. When customers are scaling up so fast, their network is the least of their concern to begin with, which is like, “Let’s just go and put our mechanical assets — and the network will follow.” So, to answer your question, we had to go on a deployment agnostic model. So, we can deploy it as a virtual machine. Now, when we go on a customer site, “Do you support Oracle VirtualBox? Or do you support Microsoft Hyper-V? Or do you support this ESXi cluster that we have?” Our answer had to be yes, and yes, because everybody brought something else. Then, there are customers, they have very non-centralized networks which is decentralized networks, which is more like, there is one router here, one router here, one router here. So, they started installing gateways, and we had to say yes, we will develop full support for any x86–64 gateways out there, Dell, HPE, and everything else. And in recent times, Kubernetes has revolutionized a whole lot of deployment models. So now, we also run as a container. We have to run as a Kubernetes Helm chart as well or a Kubernetes script itself. So, it’s just deployment agnostic completely. It can start from like hundreds of core Xeon all the way to a small Raspberry Pi type of computing power. It just took a long time for us to settle down on every deployment scenario.
Vlad Romanov: And so, if I was, say, a small manufacturer, midsize manufacturer, I come to you guys and I ask you, “Well, I would like to test this out. To see if it is a good fit for our facility. Does it involve me setting up the platform? Can I just buy any hardware that, let’s say, you recommend on your website or that runs Linux, and I can set that up myself? Or do you usually use integrators that come in from your side or from partners that come and install the system? What does the rollout process look like?”
Vatsal Shah: It’s like, we have been continuously investing in a self-serviceable infrastructure. So, customers, they can get started on their own. Just download the virtual machine and run it in the environment that they have. They don’t even have to pick up the phone to talk to us. We also introduced a live chat support recently which is, if they have a question or concern, if they have something, just go talk to somebody, and they’re going to answer your questions away. So, we are trying to make it more and more self-serviceable for the whole end-to-end scenario. So system integrators, they are there to scale up the project. Once you have one site, you want to scale it up to like 100 sites. System integrators and their humongous resources will help. For initial ones, it’s self-serviceable. And we are, like, we have some exciting announcements to make in the upcoming week as well. We are launching a central portal where customers, without even talking with us, can just go, deploy a sandbox environment, or they can download our products. They don’t have to get in touch with us. Just sign up, completely self-serviceable workflow.
Vlad Romanov: I could certainly appreciate that as an engineer. Vatsal, if I could ask you on the customer side, like, do you see them struggle with data deployments? And what I mean by that is, we talked about their networks being a little bit, “questionable”. A lot of times, I know that they’re missing, let’s say, key instrumentation. So, they have maybe a PLC on the machine, but they want to track what’s coming in and what’s coming out. So they need to do modifications. Maybe also, people are not ready to deploy something like this, so they need to purchase maybe some server hardware, which I know in some situations can be a Raspberry Pi, but usually it should be something better. So, I’m curious if you have some pushback from them or questions or concerns with what they need to do in order to get a data solution like this going.
Vatsal Shah: One of the first things that we do recommend is to understand what assets they have and what type of connectivity they provide. Whatever software does is, as soon as you open up the web user interface, it will have just one plus button. If you click that plus button, it will ask you for the IP address and the variables that you want to pull out of that control system. As long as those two variables are known, it’s pretty much a simplified workflow. For mid-sized customers, what we have seen is they still have to do some deeper discussion, or like they do have to identify what assets and what they represent, and how are they going to get access to that. For large customers, normally it’s a well-documented process, which is they already have most of the times those variables, PLC tags, IP address information. The network portion analysis is already done, so there is a little better state. What they have to prepare is to identify the assets, identify the connectivity, and the rest is on us.
Vlad Romanov: Dave, thoughts or comments?
Davd Griffith: I have to quote Vlad, I’ve got many thoughts and comments out here. But no. I guess one, this seems very exciting. Let me go back to the comment about the self-serve, self-use that you were saying, Vatsal, and I think you said that many times, you’ll have a customer, and for the single site, they will come in. They’ll download either the trial, or they’ll go buy a license and go build it themselves. I have lived on the “build it themselves” from virtually every angle. I’ve lived on it from the systems integrator coming in to say, “Hey, you guys have built something, but maybe you didn’t follow best practices.” Last year, I lived with a client as we were going through and doing a self-build with someone who, with a group that first wasn’t Litmus, but certainly did not give us kind of enough structure or anything along those lines. And I have seen kind of on the other side of people going and taking a low-code, no-code platform and going and creating some good opportunities. But I look at it and I’m like, “You guys left so much low-hanging fruit there.” So, having said all of that, I think one, being able to showcase and have that, “You guys can go roll this out yourself if you like to,” is absolutely great. And I think we will continue to see more and more companies being forced to offer those because that’s what end users want. But on that side, do you see lots of end users being able to easily leverage the Litmus platform in order to go find those big gains? Or do you find many times people come in, see that there’s a huge opportunity here, and then maybe go talk to Litmus, maybe go talk to one of the partners to come in and help them find the rest of that ROI and those opportunities?
Vatsal Shah: Sure. So, the idea would be yes. So you’re 100% right. There is a growing trend now from small teams in very large corporations which are more into digital transformation, the operations management, CIOs team, data science team. They want to be completely self-serviceable. This means they are dealing with five different vendors at the same time. They can’t pick up and talk to all five vendors when they are just initiating that project. So, they believe in the principle that everything must be documented, everything must be available to me now, and everything must be available in an environment where I can go ahead and get started. There’s an extremely growing trend on that side. And secondly, we have such a huge demand from the market. As we have limited hours in a day, our account executives also have limited hours in a day. So, why restrict ourselves? The self-serviceability is more and more opening up. We did invest in three key sides of it. The first one is distribution. So, if somebody wants to try it, it’s like central.litmus.io. We are going to announce it next week, but now I just announced it on the Manufacturing Hub podcast. But if somebody goes to central.litmus.io, sign up and you will be able to deploy a sandbox environment right now, even before this webinar and podcast is over. So that’s our distribution mechanism. We solved it for somebody who wants to just buy it, or somebody who just wants to try it for 15 days, or somebody wants us to host a virtual machine for them. We are paying it to cloud companies, but we host it for them for a sandboxing environment. So that’s distribution. The second one has to be the principle for product-led growth. This really pushes us into making the product well documented and have a customer success team ready in a way that they can help customers where they are. So, a lot of documentation. Plus, we have introduced a Solutions Marketplace in the central portal itself that allows customers to download solutions. So, if you have a funnel system and you don’t want to discover all the variables out of it, just go ahead and download the funnel template. It will collect data from CNC systems. So, the second thing that we heavily invested in is understanding horizontal and vertical specific use cases and making it live for zero dollars for everyone. So customer success and the ready availability of solutions was the second investment that we made. The third one that we are actively investing in is ensuring the product is available in a way that it guides customers throughout their entire journey — from initial connection, processing, analytics, and integration. All of these components have to be self-serviceable as well. In-product help and the improvement of user experience was another area where we heavily invested from our side as well. Combining all three of these — distribution, knowledge transfer, and product help — I believe we are ready for a self-serviceable approach. Of course, business models had to align, and we did that exercise as well. So, again that’s just over the last one and a half years, we have been investing in a more self-serviceable distribution of our product. I hope that makes sense. It’s a long way to answer it.
Dave Griffith: Absolutely. I think we might lose Vlad for the rest of the show, though, because you just told him where he could go and download a sandbox. He’s probably going to be clicking away on another screen for the rest of our time here, building and connecting to stuff behind it. I think that’s very exciting. Vlad, what are your thoughts? Do you have any thoughts on the self-service model? Do you have thoughts on other things that we want to get into before we go talk about some more solutions?
Vlad Romanov: I’m curious about data. Specifically, its contextualization once you get it into the platform. So, Vatsal, you probably know that machinery is built very differently. Even at the same manufacturer, you could have the exact same machine constructed by two different engineers, which means your tags are going to be different. There’s a noticeable lack of standards in our space. So, I’m curious. As an end user that’s looking to make sense of that data, and I hear this a lot from at least my conversations, it becomes difficult and time-consuming to first get that data. I think you’ve solved that problem. How do you make sense or maybe transform that data into something that you can then take action on? What does that look like on your side?
Vatsal Shah: Oh, for sure. And don’t even get me started on standards. They are non-existent in this world.
Vlad Romanov: I hope they will be made by someone, but I also don’t see them. They’re very different across companies.
Vatsal Shah: No, like, for sure. So, again, what we had to do was, in the 2018 timeframe, we realized that the temperature coming from PLC number one, or temperature coming from PLC number two, it’s still temperature. Everybody struggles in creating a different format of the data itself. So, we started normalizing this data at a very early time frame. Data coming across all of these different products, all of this industrial system, has to look exactly the same. That was the normalization piece that we developed almost five years ago. But, what started happening was every asset is different, as you pointed out. FANUC might have more metadata available, Siemens Cinematic might have even further data available, or a robot might have different types of metadata available. So, we created this metadata registry inside our product. Every data point that you collect, say if you collect temperature, then you can attach what batch it was, what shift it was, what asset it was, or what line it was. You can attach those metadata in a dynamic way on every data point which is collected. So, our temperature JSON message, is augmented with a whole lot of other context information at the edge. Before it hits the data lake, before it hits any other analytical systems or machine learning system, it has data which is key value timestamp plus metadata available as a part of it. This is a very basic step that every vendor needs to take. There are a lot of companies. They try to create an environment where data contextualization happens a little bit later in the pipeline, but you still lose a lot of information which was available at a plant level. Now, you cannot go back in time to acquire it in the first place. So, this is what we do at the edge, which is collect the data, normalize it, but also augment it with a lot of metadata around it.
Vlad Romanov: Interesting. When you talk about, like, a shift or batch information, can you pull that from an ERP or MES system, or do you have to, like, manually enter it? What does that look like?”
Vatsal Shah: So, yes, it can be automated. Everything that we do, we normally aim for scalability at Litmus. The activities of adding metadata or the metadata registry itself is dynamic. There is an API call. MES systems, they normally have the information on which batch, which product, which SKU it is manufacturing. We can pull that data as a part of Litmus Edge and dynamically change the metadata inside the product by making API calls. So, every time there is a shift change, there is a product change, there is a SKU change, line change, it has to be completely dynamic. And we also had to introduce a digital twins layout for that as well. So, we introduced that later last year. About 95% of our customers are using it now just because there is so much need. There is data information available, but I want to format that data in a way that I can use it in machine learning studios in the cloud, some of the modern AI algorithms in the cloud, or just visualization itself. So, yes, the digital twin was introduced to solve that challenge even further.
Vlad Romanov: Interesting. I guess the problem that I’ve seen is that, at least the companies I’ve worked with, heavily standardize on SAP, and it becomes difficult for them to make the connection between SAP and some of these newer data platforms. It becomes impossible to have an automated way to, let’s say, track a batch, track a shift, or track what’s going on. So, it really becomes a very manual process. So, it’s awesome that you guys are able to make those connections.
Vatsal Shah: SAP is one of the very highly integrated SAP PO and SAP MII. There are a lot of customers. They integrate our products to that. Their workflows are different. Sometimes they just want us to collect some context information. Sometimes they want us to push key value variables inside their environment, which is raw data. Derive, analyze data, push it in a way that they don’t have to analyze any further, and the data set is available inside the SAP MII environment or even further like the HANA database and more. But, it’s like we have to really do that, and every vendor has to do that if they want to do it for the industrial IoT world. It has to collect data from the systems which are contextualizing that data in the first place.
Vlad Romanov: Are there any interesting, maybe, I’m always curious about situations, that a customer has used your solution for something that you didn’t expect? That maybe was a little bit peculiar at first, but like, ‘Oh, now that you’ve explained it, it totally makes sense.’ Do you have any examples that you can maybe share with us? Again, obviously not mentioning customer names.
Vatsal Shah: Countless times. Again, countless times. Like, I can say that, ‘Oh, why was it not documented? Or why didn’t you tell me before? I just did it this way, and now you told me that we should have been doing it this way.’ Yes, this is the challenge that we come across almost every time. We are building the product for the first time. We are really disrupting and revolutionizing this industry. We took the product-first approach. So, we are always going to come across scenarios which are corner cases. We listen to our customers, and we try to resolve them. To highlight one of them, a lot of customers start with the easiest architecture possible. I will answer this ‘easiest architecture’ in two different ways. But the first one is, ‘I want to just push data into the cloud.’ So, what are my avenues? Which cloud vendor am I using? Am I using Azure, GCP, or AWS? Then they will find an IoT product that any of those three vendors have. One of them has IoT Hub, IoT Core, or Pub-Sub. The idea would be they start there. Quickly, they realize that it’s so expensive to push plant flow data to the cloud using those specific mechanisms. We came across a customer use case where they were spending close to $40,000 a month to push four to five lines of data to the cloud. They had a large amount of variables they were using for statistical analysis, but that is expensive. Really expensive. So, in one of these cost-reduction journeys that we found, customers discovered, ‘Oh, if I use Kafka, if I use this Google Pub-Sub environment, I might actually spend five percent of what I’m spending before or even less than that.’ Then it became a trend. ‘Why didn’t you tell us before? Why did you let us do the thing which is easy but expensive?’ So, it’s more like they discover by themselves. Once they hit the cost problem, they shift to resolve it. Us being a multi-cloud environment where we push data to everyone, that journey is like five minutes of work. That’s mainly one. Then there’s another. A lot of companies or innovators in the market start with open-source products. They think, ‘Let me just build everything by myself. I’m going to take one of these no-code local tools, I’m going to take this driver from GitHub, and I’m going to take this visualization or database and combine everything.’ Fantastic. Again, please do that. It’s the best way to get started and understand what that represents. But they soon realize they’re going to hit a wall. Who is going to maintain this four-vendor infrastructure for open-source? It’s just that they have to either invest or replicate it in an environment like Litmus Edge. When customers say they are going to build it themselves, we just mark in our Salesforce that this customer will come back a year later. Because once they realize the ROI, they’re going to return to us. It’s just a reality. We try to guide them, saying it’s fantastic to try, but once you’re ready to scale up, we’re here for you. This has been a trend. They always ask, ‘Why didn’t you tell us before about these challenges?’ ‘We keep telling you, but the message doesn’t get across.’
Vlad Romanov: I will make a comment before I let Dave jump in. There was a Reddit thread not too long ago where an engineer was asking about implementing a data, machine learning, and visualization solution on a Raspberry Pi. The question was, ‘Is this an adequate solution for the manufacturer?’ I think that this is almost like a trap for the young players in our industry. Yes, it’s going to maybe save you some dollars on the hardware, but it’s going to take you a lot longer to implement, and it’s going to be unmaintainable. So, unless, again, speaking for the engineering side, unless you want to pigeonhole yourself into just being the only person that can support that system, I really don’t see a path of you developing something that is maintainable. And again, maybe I don’t know his intentions. I don’t know if he’s trying to secure a job, but I really think that from the customer, the manufacturer point of view, that’s not something you want to find yourself in. And unless, again, maybe you don’t have the right knowledge or understanding of what that’s going to look like, you should not be signing up to such a platform.
Vatsal Shah: You’re spot on. And security, it’s the pusher that this layout creates for a security environment, is just worse. We have seen somebody reverse tunneling inside a Raspberry Pi from their home, and that data is connected to their machines and CNC systems. I was like, ‘Please do consider what you’re doing is creating a bi-directional tunnel to your plant floor from your home.’ Again, even if it’s secure, even if it’s encrypted, even if there’s complete certificate-based authentication authorization, it’s still not the wise thing to do. It’s like once again, ignoring security in favor of a lower-cost solution is never going to scale. Even if it scales, it will be unmaintainable.
Vlad Romanov: Yeah, absolutely, and I think it’s with good intention. Ultimately, they’re trying to provide value, but it’s not having the information at hand that sort of drives some of these solutions. Dave?
Dave Griffith: I was going to say, if you guys have missed, I don’t know, the middle 30 to 50 episodes on Manufacturing Hub, it was a lot of Vlad and I arguing over whether Raspberry Pi’s were production-ready. Vlad has very much the engineering concept, as you guys have heard. I am very much on the ‘no’ side. Like if I were going to go argue with internet strangers, my answer to that question would just be a ‘no’, and we would move on from there. Or we wouldn’t move on from there, but that’s why I typically don’t argue with internet strangers on Reddit because no one ever wins. Having talked about arguing with internet strangers, I’ve been having a series of very interesting conversations around digital twin. Vatsal brought up digital twin. We are going to be demonstrating digital twin at Hannover, and we’re going to talk about that. But first, we have some people to thank. So, we would like to thank Litmus for sponsoring this theme as well as this show and having Vatsal come on here. With Litmus, everyone can work from a single source of truth to improve efficiency, drive profitability, and scale securely. When it comes to industrial data, real-time connectivity, normalization, contextualization, and analysis at the edge come together in one platform to help IT, OT, and enterprises do more with their data. More than 250 drivers connect to legacy and modern industrial systems in minutes. Writing KPIs, and analytics, digital twin and machine learning models, and integration to cloud are available out of the box. If you guys are going to Hannover Messe in 2023, join me, Vatsal, and Litmus to experience ‘Real Results Within Reach’. We’re going to do daily live builds, we’ve got demonstrations, and activities to help make IIoT more accessible. Join us at Hall 17, Stand F-18 for that, and we hope to see everyone soon.
Vlad Romanov: Awesome. And I will probably join remotely on some of those streams.
Dave Griffith: Vlad will join remotely. He is excited to make it out to the mean streets of trade shows later this year. If you guys have missed it, Vlad is 100% committed on the internet to go to ‘Automate’. And we all know that everything people say on the internet is a hundred percent true. So, if you’re not going to be at Hannover, if you’re more of a North American centric, please come join us at ‘Automate’ for that show. Vatsal, having said that, we’ve got a bunch of great content coming out. I want to get maybe a couple of minute segment of your thoughts on Hannover because you’ve been there before. Maybe we’ve got some viewers and listeners who have not been to Hannover. So, can we get your kind of overview and maybe your best tips and tricks? And then we can get into digital twin a little bit.
Vatsal Shah: So, about Hannover, I think it’s one of the best shows that you will go to. It’s like, for the industrial world itself, we are seeing it changing from just a very mechanical or asset-centric show to a more digital show over the last few years. There are a lot of software vendors, a lot of talks about the modern PLM systems, a lot of talks about digital twins, a lot of talks about this industrial data ops, and more. So, yes, it has changed. Like, you’re going to be dead tired after, I bet you, Wednesday. Like, we’ll have to pull you out of bed on Thursday morning. It’s going to be that hectic. But it should be a really fun show, and there are a lot of smart people opening up a lot of different thought processes, and their innovations, and their ideas. And it’s just fantastic. The more we talk, it’s better because we are doing this live build for the first time, as Dave said. Like we are really recording things live while we are building the use case. Again, we are confident or overconfident and we don’t know. If we don’t know if the network is going to work, but we hope we are going to figure everything out. But it should be a fun challenge for ourselves. Let’s see.
Dave Griffith: Absolutely, it should be good. I will live shows again say to those live builds in some of the demos. We will do our very best to stream from the show to keep everyone up to date. A little bit of live shows and doing potentially live shows at live shows is that question as to what does internet connectivity and stability look like? In theory, we should have more than enough to be able to stream out. In reality, we will see, which is what I’ve learned about every trade show that I’ve ever been to. I will say in addition to that, tune in to this channel and the Litmus Channel. We will have daily recaps coming out as to what’s happening. And that should be very fun. They’re going to be super short. I think the concept is maybe 15 minutes max as it comes to that. Chuck, in the comments, is saying to wear very comfortable shoes. I will say one of the favorite things about international trade shows for me, Vatsal, and I’ll have to get your opinion on this, is based upon what people wear, you can 100% tell where they come from. So our European brethren, they’re going to be in dark suits, they’re going to be in white shirts, probably with club ties. If they have the company logo on something, there’s about a 75% chance it is embroidered in white so you can barely see it on their shirt. And they will be prim and proper. The Americans, and that will include me this time, we look like what we look like. You will almost certainly be able to tell us if only because we are wearing comfortable shoes and sneakers. I have decided to go the very American route of wearing like very bright sneakers. I think I love bright shirts for people to see me on, but that is one of my favorite parts of going to trade shows now.
Vatsal Shah: Well, for sure. We have to be true to ourselves. So, whatever is comfortable, that works very well. But the show is going to be fantastic, and we have a very nice set-up. We are on multiple different boots with partners as well, and the Litmus booth is also there. So, it should be fun.
Dave Griffith: Absolutely, no, I think it should be fun. Can we kind of transition a bit into Digital Twin? I know that you mentioned you guys are building digital twin to allow people to use your tools better. And as I said, I’ve had some rousing conversations as to what digital twin is. Does digital twin have to include every bit and bite? Can we simulate lines and get to 80 or 90%? Is that a digital twin? That is one of my most interesting conversations we’ve been having because just like everything in this industry, none of us can agree on it. Can you give us maybe a bit of an overview as to what the Litmus digital twin looks like and how you guys are able to go create that?
Vatsal Shah: Yep, I would say for digital twins, we divided our product management team and the engineering team. We divided into three key areas. The first one for digital twin, in our definition, is more about data itself. How do you represent data about an asset or about a specific use case? So, a digital twin can be energy monitoring for a CNC machine, or a digital twin can be the CNC machine itself. So, how do you represent that data is the first step. Organizing that data, contextualizing the data is the first piece of a digital twin. The second one is the visualization of those digital twins, which is more about 3D modeling integrated with PLM, CAD-CAM. You represent that in augmented reality use cases. You represent that in the further advanced cases where 3D simulations are involved. And the last piece of a digital twin, in our definition, is the real simulation. Where, if you manipulate a variable, how does it behave? Can you simulate how that asset is going to behave if you change a specific variable workflows. One of our large Oil and Gas customers created a digital twin of a specific asset inside the refinery. Their use case was very simple: if the viscosity of input material changes, how does it behave on the furnace on the other side? That was their specific use case. So, they hired a lot of people, they simulated in a very mathematical or physical environment by using those tools as well. And they created a simulation which was utilizing real data and which was utilizing step two, which is more 3D representation of that data itself. So, it’s a triangle in our opinion. You need to have data, you need to have a 3D representation, and you need to have simulation capabilities. If we combine all of them, it’s the most featured complete digital twin that exists now. For us, we do one and a half out of it, not even two. So, we started with an asset framework and asset models itself, start creating a very smart asset model which has a raw data hierarchy, which is ISO 95, have some machine learning model attached with that, some analytical system, some analyzed data. So, our digital twin is the definition itself. It has a lot of different variables that we started attaching, which became a smart digital twin asset model. Then we start deploying that. That is also covered as a part of the product. So, we are the source of the data. All the PLC, all the asset information, we already collected it. Now, we are just organizing the data in an asset framework environment. That’s the first piece that we did. The second one that we did was, rather than building our own 3D visualization, we integrated with various vendors who have those visualizations built out. There are two cloud vendors that have it already, and then there are a couple of other open source and closed source platforms available where we push that data in an organized format. So they can allow customers to interact with that data in a 3D or PLM environment. And the third one, not a lot of companies exist in that environment just yet for simulation because they are very purpose-built soft PLC companies. Like some other mathematical simulations or physical simulation companies, they are getting advanced in that for general-purpose simulation of asset data. So, we are partnering up with them. We are integrating with them on a day-by-day basis. And all of these three things combine. It does create a very solid digital twin framework for the customers. We are not there yet. Customers, they are not there yet, unless they’ve been very specific to their use cases. But over the next one to two years, I think the things will make sure.
Vlad Romanov: I’m curious if I could maybe ask for a description/teaser of what we can expect to see at Hannover. Like is that going to be a 3D representation? Is that going to be data? What can someone visiting the booth or maybe tuning in remotely expect of the demo? I’m really curious.
Dave Griffith: Oh, please. I know what the live builds are going to look like. I’m not 100% sure as to what the digital twin representations are going to look like. So Vatsal, you want to go walk through what that would look like, and then I can go give a little bit of a teaser as to what some of the live builds will look like, we can absolutely talk through that.
Vatsal Shah: Sure. So, inside the plan for us, we planned the Hannover message in two different segments. The first one, our existing customers or new customers, we have various demonstrations, various play environments, or sign up for our central portal environment. We are going to have all the human interactions that we are doing on one side. On the second side, which is more live build or a live demo environments, that will go throughout the day. So, there’s a small theater and everybody’s allowed to participate. It’s very interactive sessions that we are going to do. We are going to build three different use cases. Now, those are generic, and we are going to connect to the control systems or simulate them. We are going to derive the KPIs, you’re going to do the visualization, everything in a very live built environment. The idea would be, if we replicate it on a real plant floor, it will look 100% identical. Like if there is an energy-specific use case, you develop it there at Hannover in that live build environment, you take it home, download the template, and it will 100% work in your own factory environment. So, that live build is a very interactive concept. Again, we are trying it for the first time, so give us some rope and guide us if we are making some mistakes. But the idea would be, we want to really represent how to use the product to get the ROI out of the product.
Dave Griffith: And if I may dig into that a little bit longer. So, if you guys are going to be live, all of these will be in whatever time zone Hannover, Germany is. I think it’s Central European Time. So, we will have at least four lives from Hannover. I think we’re currently scheduled at about 5:30 central time. So, basically, at the end of the show, we will go live. I think the current plan is from the Litmus Booth every single time, at least Monday through Thursday. We’ll see what time Friday works with, as Vatsal I think has mentioned that the Friday is less busy and excited as Monday through Thursday because they’re going to have to peel everyone out of bed by Friday morning to show up on time. So, we will have lives Monday through Thursday at 5:30 local time. Plus, on Friday at some point, we will have a live there as well. At 11 o’clock and two o’clock tentatively, we are scheduled to have live builds in the booth. I think our goal is to bring four of those to the manufacturing hub audience live over the course of that. We’ve got some interesting builds; we’ve got some asset monitoring, we’ve got some predictive maintenance, we’ve got some process and quality control, and I think we’re working on a scalability live show. Although that is going to be very interesting to see how we showcase that. I know Vatsal has got some really interesting whiteboarding, or I think the board is going to be black, so maybe we’ll call it blackboarding, that he’s going to do on that. We will certainly try to get, or at least capture, at least a couple of those to bring those live to everyone as we go through that. So, 11 and 2 o’clock every day at the booth, and then Thursday at four o’clock local time, we will have a Manufacturing Hub live from the Litmus Booth. And then I think Vatsal is going to be on, and I think we’re going to bring up a couple of other people on it. It’s going to be kind of a panel roundtable discussion, both on creating high impact solutions, as well as maybe some of the takeaways that we have from Hannover. And again, if you guys are going to be there, please comment. Please come and say hi. We’ve got a lot of other content items that it is our goal in order to be able to create. We’ve got a lot of things that we will do our best to share during the week on socials, but we will absolutely probably have months of content that will come on after that. So be sure to tune in to all of our channels. The Manufacturing Hub channel is really good, also the Litmus Channel, and we will go ahead and have those links to the Litmus channels in there. And then, as we get closer, we will have exact channels that everything is going to go live on, so you guys, if you are not there, can absolutely go and stay tuned. I will also say, and everyone should be scared, just as Vatsal has been talking, I’ve got a list of like three or four other pieces of content that we’re going to have to create. I’m just going to shove a camera in front of his face, I’m going to start asking him questions, and done. Everyone should absolutely be worried. Yes, everyone should absolutely be worried about that.
Vatsal Shah: Like, it’s going to be exciting looking forward.
Vlad Romanov: If I can bring us back to Data Solutions for just a moment, I think we hit a very interesting point before the ad read, which is that in many instances, the end-users or the customers aren’t always aware of how to implement these solutions properly. And so, my question for you, Vatsal, is: What is the strategy for educating the end users? So, I think that the technology is certainly there, but I still think that there’s a gap in our industry of number one, like what’s possible, then what it takes, and what sort of opportunities it will provide once it gets there. So, I’m curious on your thoughts on what you’ve seen in terms of gaps of knowledge and kind of what you’re trying to do to address that and your strategy on that side.
Vatsal Shah: No, for sure, and I think technology is there. Like our product market fit, and the same goes for many other industrial companies, is very high. That means if customers are looking for a product, the products do exist. The main challenge here is that there are two key challenges. One is change management. This is one of the very heavy subjects that happens, which is manufacturing companies exist to build up, like whatever they’re manufacturing. They are manufacturing shampoo bottles, or they are manufacturing cars, or they are manufacturing books. Whatever they are manufacturing, they exist to manufacture that. Doing digitization up until this point is always a second priority for them. So, there is a change management involved without affecting their day by day operations, and that is, in my opinion, is a very delicate subject. We have to tread very lightly, in a non-disruptive and non-destructive way, introduce a technology which can help them in the short term as well as in the long term. So, change management is one. And the second one, the skill set has to grow. There are a lot of team members that we work with. They want to learn. The operational technology team members, they are ready to go to the cloud. They want to really understand how SQL Server works, or how this serverless infrastructure works. Companies are really enabling, or they have to enable them, to explore further. This transform them from just manufacturing operations or maintenance into even something further, which is the digitization of those assets itself. So, those two things which are influencing the technology adaption, and this is, in my opinion, is something that has to be taken care of as a people operation side rather than on a technology adoption side. Just my view of it. And it’s just, there’s so much noise in the market created by so many different vendors that technology is going to help them on their next billion dollar. That’s never the case. Technology plus people plus the change, done in a correct way, will help them achieve that next business milestone.
Vlad Romanov: And it’s interesting. I certainly see our industry change from that standpoint. But again, I don’t know if it’s like more different college programs. I don’t know if it’s going to be the customers that start to drive more of that change. I again, I don’t have the answer, but I certainly see a lot of gap there. Then again, it’s almost like a cliche saying that “we’ve always done it this way”, but I think there’s many — like you can break that down into what items. It’s not as black and white as saying that we’ve always done it this way. There’s many steps in educating, let’s say, your management teams, educating your operators on how to use the tools, and then not introducing everything at once because people just get overwhelmed. And I think that’s just the natural progress of things. Then there’s the skill gap, as you’ve mentioned. They’re excited to learn, but at the same time, they cannot just learn everything at once. They need to hire the right people, they need to make sure that the program is correct. Because, at the end of the day, I think when you describe the stack, and as we have these conversations, I think it’s easy to deploy, but it takes a lot of very complex problem solving and engineering skills to make sure it seamlessly comes together. So, there’s a lot to it at the end of the day. We should certainly appreciate that as a community. And I think manufacturers are starting to hire more experienced, more knowledgeable people on the cloud side, on the I.T side, on the network side. And I think, slowly but surely, it’s changing, but it’s slow.
Vatsal Shah: No, for sure. And I think it’s just, but the idea would be there is enough knowledge available. Let’s say there are enough knowledge sources out there, and cloud companies and data companies are doing a fantastic job in educating the market on what to do with the data. That might be machine learning or the modern artificial intelligence, which is opening up a whole lot of different things which are opening up. So, they are educating people that there is more to do with the data than just seeing it on a SCADA screen. And then, subject matter expertise from specific people, that “okay, if I do this by using this tool, I might be able to revolutionize our plant floor or I might be able to change this cumbersome process that I have.” It just has to be driven by people to solve a real challenge for a short term or a longer term. Again, long term does not yield results away, but they put you in a better position as a company, competitive position as well as a future-ready position, three years later, two years later. So, you have to invest in it now, and that’s what we have been saying again and again.
Dave Griffith: It is absolutely, I completely agree with that. I think the change management is hard. Anyone that knows the work that I do with end users, as opposed to all of the fun stuff that we get to talk about on here, knows that change management is a necessary part of kind of what we do. And as we look at executing opportunities, as we look at taking solutions that many times have been, in theory, delivered but we’ve never actually rolled them out to the plant floor, change management is absolutely the hardest part. But a part that we have to get better at doing and rolling out, and being able to leverage SOPs and just general procedures, and hiring the right people and making sure that they’re trained and accountable for, which is absolutely the least sexy side of this industry. But if we don’t have that, we’re never able to actually leverage a platform like Litmus or in reality, literally any platform that anyone ever wants to go use. If we don’t use it, it’s just software sitting in VMs or software living in the cloud that maybe someday someone will look at the data.
Vatsal Shah: Exactly, it has to be properly planned out. But we are seeing enough adoption in the market now. We have a large amount of Fortune 500 customers. We have a large amount of mid to large size customers where they have like 10 plants to 50 plants. They are always aiming to go one step further. How are they going to do that? They are investing in these data teams. They are investing in this industry 4.0 teams which is helping them reduce their drag, reduce their asset burn downs or maintenance schedules or the small fires that we have to put out every single day. They are using technology to avoid it so they can focus on creating the next plant rather than keep on fixing one plant. We have large amount of customers who are on that 10 to 50 plants region, and all they are doing is using technology to solve a challenge that they don’t want to solve by investing more human beings.
Dave Griffith: Yes, absolutely, I agree with this. And I think that this is a good transition to go ask you, what is the future, Vatsal. You listen to the show. You know, that I always love to get future predictions. It’s one of my favorite questions to ask because it could go in so many directions. But from your perspective, are we going to leverage technology so that we don’t have to attempt to hire more humans that we can’t find? Are we going to continue to build out digital twins? Are more end users going to become data literate? What do you imagine the future looks like?
Vatsal Shah: In knowing from customers or discussing from customers on various levels, humans getting replaced is a very low chance. Unless the things that they are doing are extremely repetitive and robotic, they will never get replaced. Plant floor, it has accumulated knowledge of the last 10 years, 15 years. AI algorithms, they can’t just come and replace a human and do something else. But there are certain things like supply chain planning or delivery scheduling, or a specific compliance report that you have to prepare, which is more robotic. Those specific items, they will be augmented by modern systems which are driven by artificial intelligence and everything else that we are seeing in the market now. And there is a very high chance for it over the next five to ten years. In a shorter term, we are seeing a continuous push where even a small manufacturer, mid-size manufacturer, they want to solve a specific challenge, and they have enough awareness that if we collect the data and if they visualize it, they might be able to solve it. So, right now, from early adapters to late size early adapters, now it is going into more mid-market segment as well. Where more and more companies are adapting digital technologies to solve their business challenges. As long as they have IT teams, as long as they have data teams, they’re all jumping on it left and right. Much faster than Fortune 500 companies, much faster. We are seeing absolutely six month projects. They are automatically getting converted into quarter long projects, and quarter long projects, they are becoming a month-long project. So, for mid-size corporations, they just don’t have legal or any other issues in between. You just want to go ahead and get started improving the project. So, in the short term, much wider adoption which goes into mid-market to a little bit larger corporations, it’s going to keep on being there. Technology, anybody who has a vaporware, they are going to get out of this market sooner or later. We are replacing a bunch of legacy systems. So, legacy repurposed systems, they are also going to get out very soon. A lot of SCADA companies, they are trying to retune their message towards industrial IoT platform. And believe me, we are doing everything we can to not let them do that. We are educating customers, we are telling that Windows-based legacy systems, they are not the best way to implement your next generation of infrastructure. You can’t maintain them, you can’t secure them. So, there is general awareness in the market what a security layout looks like and what they should be doing. So, in short term, again, a lot of those things are going to help in adoption. Long term, I’m all in for artificial intelligence. Like, there are a bunch of use cases that we are seeing. Nothing concrete yet and nothing business critical yet, but it’s getting there. The systems are getting smarter every day.
Dave Griffith: Oh, fantastic! I think that those are some amazing predictions. I hope that at least half of them come true and continue to come true because that will be good for all of us in this industry. So, thank you for that, Vatsal. The next question I like to ask everyone is for a book and/or content recommendation. And I think when I asked you this question, you’re like, “I have a number of each.” So, would you mind kind of giving us a book recommendation and then if you’ve got some other pieces of content recommendation? We always appreciate those.
Vatsal Shah: So, the book that I’m reading now is by Robert Iger, Disney’s CEO, “The Ride of a Lifetime.” It’s one of the fantastic books that I have read. It’s ongoing, I’m like 75% there in finishing it up. So, that’s one of the books that I really like. The recent one, like a couple of weeks ago, I was on “A High Growth Framework” by Elad Gil, which is also a very nice book. Mainly for the companies who are scaling up from just getting started to the next big milestone that they want to achieve. There are enough successful stories there. So, both of those books, I highly recommend. One of them is very business specific and the other one is about startups and how the stream works. As for content itself, surprisingly, I have subscribed to 10 different Substacks now. Like, seven of them are related to aggregation of artificial intelligence as well as the business in general which are going on. They send me one newsletter every day. But, there are a lot of new media and content creators which are out there now, and they are doing a fantastic job in aggregating a whole lot of content, putting their own spin towards that message itself, and then making it available to subscribers like us. The idea would be, I just don’t have enough time in my day to catch up with all of them, but at least I’m subscribed now.
Dave Griffith: Okay, I love that! I think all of those are fantastic. Thank you for that. So, everyone knows the next question I like to ask is for some career advice. So, Vatsal, perhaps we can put a different spin on this. We’d love some advice, some career advice maybe for someone who is where you were 10 years ago, considering building what has now become Litmus. Do you have some good advice for those people, early to mid-career, saying “Hey, I can build something better” or “I have a new novel idea”? Other than coming to work for Litmus, because you guys are obviously the best, what would be your best piece of career advice? I think when you’re building something for the first time, it’s just incredibly difficult. I applaud all the startup founders out there who are starting something for the first time. But, on top of it, if you are starting in the industrial world, it’s even more difficult because the industry is ruled by large companies, large distributors, large system integrators, large vendors. So, to make your mark, you have to persevere through it. There are going to be lots of downs before there is one up. And if you’re building something in the industrial world, if you have a novel idea, do try it out. The industry is prime for innovation. It’s ready to adapt to new vendors if they are ready to try out new things. It will just take a little bit longer than you anticipate. So, you do have to put hard work, smart work, and persevere out of it. It’s just my advice. I like how I started again, from where we are now, a lot of smart people joined us, much smarter than me, while building this company and we jointly grouped together. So, we have to persevere. We are in for a long run rather than there is not going to be a quick success in industrial or a world where critical ecosystem relies on you.
Dave Griffith: So, thank you for that. I think that that’s an amazing message that a number of us probably needed to hear, and maybe we’ll go clip that and just replay it every morning for the alarm clock because absolutely, it is difficult. It is especially difficult in this industry, but once you find product-market fit, to your point, the sky is the limit. So, thank you for that. The last question I like to ask everyone is, who should reach out? How can our community help you and what you’re doing? Are you guys hiring? Other than coming to visit us at Hannover and the other trade shows Litmus is going to be, are you guys looking for new customers? Kind of an open platform for you.
Vatsal Shah: So, if anybody would like to try our product, go to central.litmus.io. This is the public website. Sign up, deploy your sandbox environment or download the version of our product. Now, that’s how we are available, we are open, we are transparent, a hundred percent there. How do you approach us? Feel free to explore the Litmus website. There are a lot of resources available or we are happy to walk you through anything that you are looking for. That’s on the second one. We are hiring. We are hiring across the board, anywhere from engineering to data science teams, all the way to solutions architect, and customer-facing teams, customer success, and application and network teams as well. So, yes, we are growing, and we are very much looking for smart people. We are very much looking for young talent, experienced talent, anybody in between. We are hiring all across the board. So, please do apply. And if you cannot apply for the position, just send us a message directly and we are going to get in touch with you as soon as we can. And of course, we are always looking for more customers to join with us in our journey. So, happy to discuss. I’ll join the first call. If you come out of this specific session, I’ll be there. That’s the first session, I will join you.
Dave Griffith: That is a very interesting offer. Thank you so much. Thank you so much for that, that’s all. Either it will be a good offer for a couple of people or you will come to regret that offer.
Vatsal Shah: I’ll have to live with that.
Dave Griffith: Yes, but no, so thank you, Vatsal. Thank you, everyone, for coming and joining us at this slightly earlier time. I would like to say that if you guys like this time, if it works for you, please go ahead and drop us a message. We always like to hear different comments about different times as we continue to trial it out. Being a live show, as everyone who watches us live knows, sometimes noon is great, sometimes four o’clock is great, but there’s at least 40 to 75 percent of our audience that cannot make it at any given time. So, please let us know. I would say again, thank you to Litmus. If you guys are going to be at Hannover Messe 2023, come visit Litmus, come visit myself. We again are in Hall 17, Booth F-18. We are right next to Google. So, I mean, I would say you couldn’t miss us, but I think it’s so huge you probably could miss us. There’s an app. Go star Litmus on the app and come visit us at some point. Absolutely important, that should be a lot of fun. If you guys are watching live, please remember to hit that like button. Please go hit subscribe onto the Solis PLC YouTube channel. Please like and follow Manufacturing Hub Network. We are almost to 2,000 subscribers on YouTube, which is pretty awesome, mostly because we never remember to ask anyone to subscribe to that. If you guys are watching, Vatsal’s LinkedIn, myself, and Vlad, all of our LinkedIn are on the events themselves. You guys can please feel free to connect with us. Go to central.litmus.io to go test an, as of yet, a completely open demo that you guys can go on. Again, either we’re going to be lauded for getting some early users or going to get yelled at shortly after this, Vatsal. But that is the fun of a live show. And if you guys have made it all the way to the end, listening on podcast forum, please hit that like. Please rate us five stars and follow along. It helps the algorithm, and I’ve learned if I remember to ask, you guys remember to do it. Until next time, we’ll see everyone soon.
