Sequoia Leads $75M Series B Into Nominal | Alfred Lin Joins Board

32:20And so a lot of the testing activity at SpaceX kind of falls into this general bucket of Delta qualification and how quickly you can provision that test, collect the data, analyze it, build that body of evidence to go present back to your customers, your regulators. That is what it means to do continuous tests and continuous validation and to do iterative development in an agile way on a hardware product. And SpaceX is very, very good at that. And we're helping a lot of other companies become very good at it as well. delta quoll delta quoll yeah it's a big i mean i'll take like i think it's the thing we see um you know steven was sort of talking about it in the commercial context but um it's this is a i think a huge area for the dod broadly as well and what happens often is i mean these programs are massive really big budgets very you know very very costly from just the you know budget perspective but also the human beings that go into these, you know, these programs.

33:18And so being able to help, you know, in the DoD setup, like it has services, right? And so it's like, there's intentional ways of, you know, the Navy version of the F-35 versus the, you know, Air Force version, right? What we see is like, there's a lot of retests that happens that probably doesn't need to happen, right? And so this concept of DeltaQual, like our vision would be to be able to provide the tools to determine exactly what does not need to be retested. You know, if I'm doing the variance between the Navy and the Marine Corps for the F-35, for example, like very small, some significant differences, obviously, in what the operating picture and like what the vehicle is supposed to do.

33:59But a big swath of that is the same and you shouldn't need to like retest a lot of that. But often that's what's happening. So to get deeper into the products, can we start off with the unified operating platform? yeah absolutely we have two products that we'll talk about today uh and then there are some more in the works but um the two products that are out kind of on the market now nominal core that's what we started with um i'll talk a little bit about like what how that you know got got off the ground um and then we're really excited we released um just honestly a month or two ago uh nominal connect which is our second product which has been really exciting um and we can dig into that a little bit But I think it's just telling the story of where we got started.

34:43We got started specifically building a product that would... This testing and validating hardware is a massive process. Very, very complex. We wanted to do a single thing really, really well. And so that was this data review process. So if you're an engineer and you're building hardware, everyone knows what that term means. It gives you an emotional connotation of just like, oh, God, I have to do data review. But making it just really easy for engineers, like after a test has happened, to collaboratively interrogate their data, have it organized, make sense of it, do both like ad hoc exploration.

35:17So like find data, query channels, just like have a really interactive picture of their data. And then also start to build out sort of validation logic and rules that can automate that process because it's highly manually intensive. Like a thing we hear a lot is you have really expensive engineers, very smart engineers that with the way the software is sort of the status quo, they're spending a lot of time doing like very manual processes of their data. Things that can be done by software can be done by automation. But it requires a really powerful interface for those humans to be able to sort of like enrich that data and make sure that it makes sense.

36:02So we started with data review and the product has obviously grown. This is, you know, that was three years ago now has grown extensively since then into this like beautiful product that is nominal core. Stephen, I'll kick it to you to give the sort of core overview and then talk about Connect. Yeah, it's, you know, like you said, really a unified platform. And the way that I generally break it up when talking about it is these three main workflows. the first being data management, second is data analysis, right, more of a manual process, and then the third is the automated validation. And you kind of want that second piece to flow into the third over time, where as you're learning more about your system and doing that analysis in nominal core, that you're starting to move those learnings into the rule set to automatically evaluate it.

36:51The data management side, though, is very, very important. And it's something that ironically is, you know, we want it to be invisible to the average user. We don't want a nominal user to be thinking about what is the particular file I need to look at today? Where is it in my file system? What's the right version of the tool to go and analyze that? And so when you look at the makeup of our engineering team, about half of the headcount is backend software engineers and just brilliant people from Palantir, from financial industry, from big data, folks that are used to these really heterogeneous complex data problems.

37:25And so when we say data management, it's sort of two pieces. One is how do we go and be able to upload or ingest all of the relevant data into our platform, whether that's sensor data or video data, high rate, low rate, regardless of the source, be able to pull that in and do that very, very quickly. And we're very opinionated about meeting our customers where we are. And so rather than just kind of throwing them some documentation and saying, we do have a really good documentation, it's beautiful. But rather than just throwing them some documentation and saying, figure it out, we're actually going and building what we call these first class integrations with these data types.

38:02And this is one of the benefits of being focused on hardware testing is that that's not a super long list. We can go and build out all the necessary file upload routines, streaming data connections that we see across these industries so that we can get running, like I said before, within hours on day one of starting an account. And then the second big piece is presenting that data back to the user and doing it performantly. The performantly piece is very hard, hundreds, thousands of nominal engineering hours into ensuring our backend storage and compute systems are highly optimized, highly scalable.

38:36But then as a user looking into that data, it should be presented to me in a way that is representative of what was happening in the field. And so we think about runs being these test events. And so a lot of nominal customers, that just enters their vernacular where they're talking about nominal runs and they're referencing tests that are happening on the production floor or flight tests or customer operations. So the run is kind of the real thing that happened, bound in time. And then the asset, like Cameron mentioned, the asset is the actual physical device producing the data. That could be an entire aircraft, or you can make it as granular as a single motor or a single circuit board on your system.

39:13And so that data model maps very well to the way that your average engineers are going to think about their systems and what they're trying to evaluate. The data analysis piece, you know, briefly, it's really just kind of merging these Grafana and MATLAB-esque workflows together, where Grafana is a very good tool for extremely agnostic cloud data and really just looking at time series data. And so at a company like Andrel, very early on, they might implement Grafana for fleet monitoring, right? And then you have MATLAB at the other end of the spectrum, which is very, very engineering focused, very hardware focused, that has all these toolboxes for different specialized analyses, but is locally deployed, right?

39:53I'm running it on my personal laptop. And if I want to share the results with somebody, I can't send them a link. I need to take a screenshot of the plot or send them 4 ,000 lines of MATLAB code to run. So we're lifting both of those workflows into core, into the analysis platform, so that people can do this collaborative, really deep data analysis all in one place right next to where their data is stored. And then the last piece, like Cameron mentioned, is the automated verification. And so whether that's validating that a particular component performed well on the production line or running just a continuous evaluation of data coming off of a fleet of assets, you can create those rules in nominal using what you learned in the analysis platform and deploy it everywhere.

40:34And so that's the unification piece there, whether it's production or simulation or customer operations, consolidating that single source of truth for what write looks like in your data. What about Connect? Connect is extremely exciting and close to my heart because my work at SpaceX and my work at Cerespace before that was very, very close to the hardware, right at the edge, as people would say. And the things that make Connect really special, the first is this is a desktop native application. Web frameworks are wonderful. Core is a fantastic performance scalable product. But when you are out in a propulsion test facility and you need to be able to abort a test within five milliseconds if something is going awry and you need to go into a safe condition, a web framework is generally not the architecture that you want for that.

41:28Traditionally, you're looking at something entirely custom or you're calling in a third-party provider to implement a PLC-based architecture. So there's this huge market for deploying an application at the edge. And Connect has been just a fantastic rollout over the last three months and a fantastic time to value. And really what it is under the hood, right, you have this very, very performant data backplane for pulling in data. But this can look a little bit different than core's lens into data, right? Connect, you're thinking more about PLC pub sub networks, extremely high rate, potentially frequencies up into the megahertz.

42:08So pulling that data in, organizing it, again, presenting it back to the user, and doing that in a commoditized way. So we're building out all of these different drivers for different hardware systems, whether it's national instruments or more hobby systems like LabJax, for example, or PLC protocols to be able to plug in quickly. And then on top of that, we have this great sort of widget-based architecture for running these tests. Which is based architecture. Wow. And that's everything from I need to run a checklist to I just need to look at some data on a plot to I want to actually upload data to core, to the mothership, as we would say.

42:46The mothership. And so between this driver layer that's connecting into the hardware and then all of these very easy to use GUI widgets, we can walk into a production line or into a test facility and our connect team, our instrumentation engineers, can get to value in less than a day, right? They're kind of looking at this smorgasbord here of different things to pull in from our Connect framework, implementing it, and then giving it to the customer such that they can, if they want to, continue supporting it themselves. We're not giving them this black box of like spaghetti lab view code, which is the case for a lot of these solutions.

43:22We're giving them something that's more modern, more transparent, and so that, you know, they can help extend and build it if they wish as well. I'll like I'm just super excited about connect so I'll hit on a couple things too which is um so connects being like you know kind of run and managed by Jack Palmer who we hired uh almost a year ago now um but Jack most people know Jack as the former founder and CEO of Plotly so one of the largest open source you know visualization um libraries and kind of companies extremely successful is used by many, many customers in our industry. But Jack built that business, ran it for a decade, got back to his roots and was sort of building software in the hardware world for a little bit.

44:09And we kind of came across paths and decided to join forces. We hired Jack. And as Steven said, we broke ground on, kind of got around the table as a executive team and sort of Jack had this really strong thesis that this product needed to exist. we'd heard a lot of pull from customers um and slap the table in in december and you know in january it was uh you know mvp'd and was already like selling on contracts right and so like that that was really exciting for us um it's now you know doing really well and we're we're sort of doubling down on um on that team we're here yeah we nominal just raised a series b but i always talk to jack like we kind of made a seed bet for connect and like now i want to preempt the a it's like Jack go like go really go really scale this because it's it's working really well we often get asked you know a question investors would ask or sometimes people just ask in general they're like you know you've got this infinite tan but like how generalizable can you make a product like nominal to serve all these use cases like how could it be useful for someone that's doing flight testing and also someone that's just doing manufacturing and then someone that has building a nuclear reactor, like a few fusion, all of those are nominal examples of nominal customers.

45:23Connect is kind of like our demonstrative answer to that, which is by actually building an application framework. Like we have very sophisticated, very technical users that are proficient in Python and can build a lot of this like last mile work themselves. And so I think it's been like super powerful to see our users in addition to sort of our instrumentation team start to build up these applications in in our framework um and just do that last mile to get um connect valuable for them but then also to get data to the mothership as we um as we kind of said and then the third the third thing is uh connect is exciting because it's also kind of like a direct um replacement for existing spend categories so my my business hat on um national instruments labview siemens you know massive software industrial software conglomerates that have had you know billion dollar businesses here for for decades um generally there's like a very low nps for those products like people don't like them um they're not very good uh and you know core is this like vision it's sort of this light like on the hill like you know shining light on the hill that's like here's a better way of doing organizational test data management and hardware data management for your organization.

46:45And Connect is sort of this extremely practical, you are already using a similar tool like this right now. Nominal is cheaper, better, faster. You already have the line item approved in your budget. You should just swap out. And so it's helped us really go fast from a business perspective there too. And I think Stephen made a point earlier about the supply chain broadly. It's been really cool to work with customers that have one of the first questions or second questions they often ask us when they're using Connect is, you know, oh, my God, like I have a supplier who I work with across the country.

47:25And there may be a 200 person company and we rely on them. They make a widget or a component or a servo or a pump. And like, I need to interact with that data. Can you extend a nominal license to them? And so we've done that model where we've sort of given Nominal Connect to those companies, companies that we just naturally would not have probably come across unless we were actively trying to seek those companies. So there's this huge industrial, I'd say almost middle market that I think Connect is helping us sort of break into in a really exciting way. So needless to say, we're bullish on it.

48:02But actually talking about spending smarter and moving faster. Sorcery is sponsored by Brex, and they're all about performance. So in this next question, I want to ask about performance. So you're entering a legacy industry, one that's super complex. Each scenario, each case that you're working on is unique. How do you guys manage performance for your customers? Yeah, I think I'll hit it from a couple. of different angles. The first is, and maybe pushing back a little bit on the uniqueness, one of the things that's made nominal possible now, or say three years ago, is that there is this crunch happening across all of these disparate industries in terms of the data types that they are using.

48:50Where 20 years ago, if we were going in and trying to integrate nominal, either connect or core at N different companies, we'd be looking at N different potential data integrations, because everything was so bespoke. But as everybody's migrating to more common data models and you have also this diaspora of SpaceX and Anduril engineers going out into the world and bringing these best practices to industries like nuclear, like maritime, we can go and plug into these, you know, previously fairly unique workflows in an equitable amount of time. And then everything downstream from that also looks quite similar.

49:26The data management problems are the same. The analysis problems are largely the same across industries. And we need to be a little bit focused. And if you read some of our previous media, we've been very focused on flight tests, for example, and recently posted about test facilities as well. And so we are kind of marching through and thinking about for the analysis workflows, for visualization, let's make sure we really nail these one at a time as we move through the different industries. But from a performance standpoint, the additional thing I'll add is there's a lot of really fantastic open source or even licensed tooling that's available for us to leverage.

50:03And so, again, 10 years ago, if we were building this platform, everything would be from scratch. There just was not the level of performance in the openly available systems to go and integrate. But we can do our own sort of buy versus build decision as we're looking in and saying, hey, we need to set up this particular functionality in our cloud stack or in Connect. Is there something out there that we can pull in and leverage and build upon ourselves? And so we're doing that and making sure that we're using these industry well-known, best-in-class tools to pull in. But then simultaneously, we're building an immense amount of custom code and just the nominal proprietary magic sauce.

50:43And a lot of that has to do with horizontal and vertical scalability. And if we bring in a system that works very well in isolation, we might need to do additional engineering work to ensure that we can scale it to meet either surge demand from one customer or the introduction of 10 times or 100 times more nominal customers over time. but then also just continuously testing our own stack, right? We don't want to wait for a customer to push an immense volume of data to us to discover that our ingestion or compute is insufficient. And so we're getting ahead of that, ideally months in advance and doing synthetic testing of our stack, pushing representative data loads through it and just exercising it both discreetly as well as continuously as we're making code changes so that we are not keeping up with our customers, but we are well ahead of where they are in their data scales.

51:35And generally, I think we've been very successful. You're not secretly building nuclear reactors in the back just to test it? No comment. No comment. We do have a really wonderful hardware lab. We posted, I think, about it at some point. But we've bought all of this great data acquisition hardware, these different sensors. And so we have our own, we call HIDL, hardware in the loop lab, where we are continuously streaming data into our platform from systems that are representative of what our customers are doing, which A, is great for testing, but B, is really valuable for our own engineers to kind of get hands-on with a lot of this hardware.

52:09So I do want to go into some case studies because I think this will be so fun, actually applying it to some of your customers. Maybe if we could lay out three different kinds of industries and how nominal is being applied to those specific problems or solutions. Yeah, awesome. There's three great ones we can talk about and have some public case studies on too, but we can dig even deeper here. I think one is, you know, aerospace, traditional. We talk a lot about flight testing, but we, a customer we love and have worked with for a while now, Shield AI. So we can dig in there. I think another one, Stephen kind of referenced and you talk about, we're not building a nuclear reactor, but we're trying to help people.

52:48And so a company we work with is Antares, which we love. And then a third one, we talked about kind of the multi-domain aspect, but VADEN systems. builds submarines, small UUVs, near and dear to me, obviously. That was just a personal one that I love. But no, I'll start by talking about S.H.I.E.L.D. for a bit and then see what you can cover and we'll kind of go through. But yeah, S.H.I.E.L.D. is really exciting. It's been over a year now for the partnership. We came across at a really exciting time where they had actually just brought in a new, I think then, you know, sort of VP of hardware engineering, but this gentleman by the name of Armour Harris, very well known, I think, in the SpaceX kind of community, but he's coming over from SpaceX.

53:31I think he's now the senior vice president of aircraft, so all of the aircraft sort of development for S.H.I.E.L.D. But he came over, I think he had experienced the vertical integration benefits that SpaceX had invested in. And so we hit him in a moment where he was coming in and I think rethinking and thinking about how S.H.I.E.L.D. was going to sort of move into this next moment for the VBAT, which is their vertical takeoff and landing drone, really awesome product. and I think Shield had sort of won some big and exciting contracts for themselves. And so they're thinking about, you know, how are we going to go through all of the testing, kind of Steven's like DeltaQuil, like how are you going to go through that exercise as you produce new variations, new variants of this VBAT drone and try and just do it as fast as possible.

54:15It's been really exciting. You know, they, Nominal has sort of, I think become kind of the de facto piece of software for experimental flight tests, developmental flight tests, how they're going about iterating on the VBAT. And it's been really exciting to go from a mode where I think just given some of the software tools that were available to them, they were using a lot of Python scripts and MATLAB and then sort of bridging into this world where you have nominal, where everything's really collaborative. And you can just get kind of cross-functional engineers that mechanisms or avionics or GNC or flight software or even more just like mechanical engineers access to flight data that was just very arduous for them to kind of get access to before.

55:03And what that does from an organizational perspective is like it just allows you to test faster. And so with them, it's been cool to work through. And this is very much a it's less like customer. It's like an actual kind of partnership. And they've been really awesome in pulling our product development forward. But the outcome we're able to deliver, like why I think they're, you know, been excited about nominal is testing like more tests a day. And when you think about that from a six, nine, 12 month, you know, program development lifecycle, that's when you can take a step back and be like, what if, you know, using nominal and investing in nominal as an organization, instead of taking nine months to get through the development and validation, it was eight months or it was seven months.

55:44Like, what does that do to your top line and bottom line as a business like Shield? That's been really, yeah, that's been a really exciting one that we, yeah, we love the Shield team. It's been great. So you want to talk about, yeah, Antares and Madden? Yeah. Antares was an early partner of Nominal, one of our earliest customers. And I think, you know, they, as we do, saw the light very quickly that this was a tool that they wanted to put in place early. Right. And we'll often hear folks arguing for, oh, we'll wait a year, we'll wait six months. Rarely the right decisions. You want to get these platforms in place as soon as possible.

56:24And Antares took that bet and we rolled out the platform and have been kind of growing with them, which is very exciting. We had a fully functional and working product when we started with them. But as they're scaling their production of modular reactors, nuclear modular reactors, and we're scaling our product and introducing Kinect, they've been this fantastic partner and really just a thought partner for building out features, getting feedback. And I think one of the major pieces is just the usability. And one of the hard things when you're hiring very quickly as a small company is as new folks are coming on board, you need to onboard them to all of your systems.

57:02And for most engineers, those are analysis tools, data management tools. And one of the big values that we provide to all of our customers, but it's most apparent, I think, at these earlier stages is they're trying to get people in and being able to do analysis and deliver value as discipline engineers as quickly as possible. And if you're spending two weeks learning about the tool chain and learning about where to find the right data, that is a ton of lost time when you're a 10 person or 20 person team. And so as we've seen their group growing, bringing folks into the platform and seeing our focus on usability really just kind of coming to fruition or manifesting here as they log in, they start doing work.

57:45Maybe they have a question or two and they're just going, right? There's no like onerous training thing that we have to go through as they bring in new folks. So it's been cool to kind of scale alongside of them. And like I said, they've just been really, really great partners. And then on the maritime side with Vatten, similar, smaller firm scaling up. One of the things that was really interesting there was, again, the speed that we were delivering value. And doing that roughly six months after Antares. Antares was maybe a year ago and Vatten started eight months ago or so. And being able to integrate their data quickly, get those pipelines up in the matter of a few days.

58:28and they lifted and shifted their analysis workflows into our platform very, very fast and just kind of did this cutover. And it was a moment for me looking at this and realizing like this is going to scale and we are breaking free of this world of a multi-week or multi-month implementation challenge that you see in a lot of other really heavyweight platforms. And the other piece that was interesting with Fatten, well, A, their product is awesome and I got to go out to their facility and meet the team and go out on the boat, which is very cool. But what we're seeing, you know, maritime, nuclear, aerospace, is that we're not getting bombarded with a million different maritime-specific feature requests, right?

59:08We're seeing a lot of commonality across these industries. Folks want the same answers. They want to have the same processes. And so in a very good way, right, they are using the product that we have built for everyone. And, you know, they'll, for example, Vatten came in and said, hey, we need to have bathymetric maps. when we're looking at our data, looking at the location of our assets, it's not very helpful to look at a satellite map of the ocean. We want to actually see the bathymetry and what are the contours of the earth underneath the water. And so that we went and built and rolled out, and it's benefited our other maritime customers as well.

59:41So there's these little unique things that we need to do. But in general, the universality of the telemetry management analysis world has just been very exciting for us to build upon. How much time and cost do you think you save your customers? Yeah, it's a good question. As the company has gotten more mature over the last three years, I kind of think about, there's three value conversations that we get to have. One is often the easiest, which is a quote I'll often hear is like, oh yeah, we hired a really good software engineer to build our internal version of Nominal. and then like, whoa, I didn't realize like two years passed and there's now like 15 people on that team.

1:00:28And so we hear that time and time again as companies, you know, scale. And it just like, there's a lot of overhead that goes into maintaining a productionized quality version of something like nominal. I say like nominal because they won't be as good. But it's costly nonetheless. And so I think like we can often have a conversation there which is just like how valuable is it for you You did not have to staff up to four or five, six people. It's like minimum a million in overhead. Yeah, exactly. Fully burdened, you know, engineering employee. Like, and so that is the value conversation we have for sure.

1:01:02I think like the next layer up is probably that sort of like incremental time savings. And so if you're, you take a flight testing customer and you can say like, in a given day, you can test six times when prior you were testing three times, like three times a day. And often it's like this is a human. Sometimes it's literally like a bus full of humans that are bussing like three hours away to a test site or a test range, you know, performing test operations like think about the salaries and all the ops overhead for like those humans. And when you like the moment when they come back and realize like, oh, that one test like didn't actually get the data we needed to or like those types of incurrences.

1:01:42So there's like all of that just efficiency spend of what we're delivering for folks often, you know hearing things like uh things that used to take 24 hours taking like under 10 minutes right and so like when you stack up those processes in an individual engineering layer like it used to take me 24 hours to just get to look at this piece of data and now it's 10 minutes and then i'll go back to like the stuff that's exciting for me is then when you actually get to go to like a executive or a vice president and be like um what if you had confidence that you maybe could like beat your competition because you're moving faster on the like month time horizon or you could increase even incrementally by a small number of percentage points the probability that your system is going to work um when it needs to on a tight timeline those are uh those are i think value discussions that are measured in the like percent of contract and so if we go to a program and they say, hey, we just want a 10 or a 50 or a$300 million program.

1:02:46And Nominal is able to sort of make the case that if you invest alongside us as a partner, you're having a very exciting, I'd say, value conversation at that point. So you guys are creating immense efficiencies for your customers. We're in a rapidly changing AI world and environment. And alongside that, you're growing a company. So how do you manage innovation internally with these huge macro factors? Yeah, I mean, I'll give a I'll kick it to you, Stephen, for some of the specifics. But I'll give I think a general philosophy that we have at Nominal is like always thinking short, mid and long term.

1:03:26And a lot of companies say that. But like we one of our core company values is cultivated advantage. We talk about it all the time. And so I think we know in a nominal, this isn't a quick adventure for anyone. Like we want to, we're building a generational company in this space and it takes time. And so I think we think and invest in long-termism whenever we have the opportunity to. So I think what that means is like, we look at this space of software for hardware broadly. I think of it as one where, you know, AI has not yet taken a foot. and I honestly think it's going to be a little bit of time before it really does.

1:04:05There's a lot of, on the government side specifically, I think there's been a lot of like false goods that have been sold, predictive maintenance, prognostic, you know, type things that when you actually get down to it, like the data, the telemetry and the sensor data and the logs, like it's so unstructured and so messy and so disorganized that it's not really possible to actually, you know, do proper you know machine learning or just train inference in general on on that data so we see nominal as actually a necessary like stepping stone um there uh and so i think like you know that's that's a big area um but i do i do get excited as i look at more um like agentic workflows i think that's an area where um the workflow of like what an analyst or a test engineer goes through when Steven has gone through this like thousands of times in his career, like looking at test data and going through that process.

1:05:01Like, I think there's a lot of power you actually can prompt humans with to help make their life easier. Once you sort of build this, like, this base of like what it's like to go through that workflow as a test engineer. So I just say we're far from being anti-AI and we're actually, I think we're working on some exciting stuff there, But we think that it's definitely a crawl, walk, run kind of process. You want to add anything? Yeah, I think you nailed it. The introduction of AI into a lot of these robotic systems is itself a market for us to be able to go and validate what a lot of people see as black boxes.

1:05:38And if you were to go on top of that and say the nominal validation software is also leveraging AI to go and evaluate the AI-enabled robotic systems, you start to probably raise some eyebrows here. Like at some point you need the human. It's become self-aware. The human tranche has to be doing the due diligence. Testing itself. Yeah, yeah. But there's good opportunities for us to bring that into our own platform. Things like natural language queries, things like anomaly identification. And the latter, you know, to me is more like the, it's like the clippy paperclip in Microsoft Word, right? It's not doing it for you necessarily in the analysis workflow.

1:06:11But it's saying, hey, something looks odd here. You should go check it out. Here's why I think it looks odd. And then there's still a human in the loop to do that actual due diligence. i'm curious because you have such a wide set of customers what exactly is your go-to-market strategy across them and are you your dual use or yeah yeah we're definitely we're definitely dual use um yeah it's on that point uh in the dual use nature like i think um we've been really excited like this the exact same product serves commercial and federal customers so there's like zero differentiation with either core connect there it's been a big like north star of like how do you not drive yourself crazy from a resourcing perspective?

1:06:51So like the same product works there. The go to markets are totally different, obviously. And we, I think, have been fortunate to invest in just like a killer federal BD team, folks that are very experienced. Like that's an area that I'm super passionate about and and have a good, good depth of experience. And then on the commercial side, yeah, I think just getting folks that span the, you know, span the gamut of like very technical sales And so we have folks that have sold, you know, databasing technologies and like they they love the technical cell. And honestly, a lot of it we see is like, if you read about the challenger cell, like a lot of it is sort of, hey, why?

1:07:33Like, hey, you know, company that's doing this, like, why can't why can't you do this better? Why can't you do this faster? But done not in an antagonistic way, but done in a collaborative way. Like there is a better way. Like, let's show you and let's work together to get there. That opens a lot of the doors. and it's totally different again if you're serving a legacy prime where you know it's it's how can you navigate 15 phone calls before you like get to a decision maker that has like some you know some input or some you know some authority so you you add anything in that i think it's great yeah i think like um it's it's hard like we're setting up multiple go-to markets uh and kind of routines at once.

1:08:17But, you know, I think there's, there's nothing. It's very simple. It all kind of boils down to, I think, like pincer movement, that sounds very aggressive. But I think it's just like building the technical alignment and base case with end users is like a thing you always have to do. And then being able to have this sort of like executive level alignment over what what values actually being delivered you know if something like nominal is adopted that is true across all the instances so yeah i'd love to go deeper into the nominal team you guys have attracted tremendous talent from really big names so how many employees do you have and where are they coming from yeah right now we're uh we're about 60 a little over 60 growing pretty rapidly um we are hiring check out um there's jobs on the website and they'll probably be posted in this thread as well uh we're hiring for uh pretty much pretty much everything um hiring for everything hiring for everything a role that is like very uh we love and we can't get enough of is mission operations and so that's uh uh for us that means it's kind of like one-third account manager for our you know for our customers one-third like product management um and i think one third like technical sales.

1:09:38And so it's very exciting. We get we have folks who came from robotics backgrounds, folks who did flight testing, you know, folks who were systems engineers at NASA and also just like love engaging and talking with customers. And on the pitch, I can say like in a single day, you might be able to work with a company like Shield AI and Antares and Vatten, which is pretty sweet. You could be out on a boat like cruising around with UVs and looking at a new generation small, you know, small modular reactor and watching autonomous, you know, VBAT fly around. It's like, it's pretty, it's a pretty sick job.

1:10:13So we're hiring, we'll probably grow to, I'd say probably over a hundred this year, this calendar year, to just meet, you know, meet customer demand. And then, yeah, I think like the secret, the secret sauce or a lot of what we think about is so much of nominal or really world-class software engineers from diverse backgrounds and then there's this perfect ratio of the like steven archetype um so someone who spent a lot of time operating hardware um because the beauty comes and we talked about we have this little like hardware lab like the beauty comes when you can expose these amazing software engineers to the like the brutal world of where software meets hardware um and just people who've done it who've been at you know disconnected environments at test ranges all over the world um and so that's been like a big part of the nominal culture is like getting all those folks under uh under uh under one roof but yeah it's team is a big a big thing and i think it's uh it's definitely one of the things like i personally am most most proud of it what we did have assembled a nominal and it's just it's just getting started so what about the existing team any shout outs um i feel like i've given you know steven uh shout outs but um yeah Stephen you want to give some?

1:11:30I think I reflect back on some of the initial folks that I met who are all still at Nominal and in very senior positions now leading the team these really fantastic engineers from Palantir Helen, Mark, Ross Andrew, just this great initial cohort of absolute hitters when you talk about talent I mean there's like the real recognizes real moment when people come in and they meet these folks and And them having spent between five years and a decade dealing with really complex data problems has just been absolutely immensely valuable for us. And then we're supplementing those backgrounds with folks from a dozen other different industries, two dozen different companies.

1:12:14And you get this really great mishmash of different lenses into these problems. And so I think about our folks from Applied Intuition, Nick and Drake coming from the autonomy side from a very customer focused lens, excellent engineers, Michael from Relativity Space, standing up his own data architectures there. so it it it continues to floor me that every new person we bring in is just up leveling us in a new way bringing new talents and it's just an incredibly smart and a team that just works together very well it's very it's very easy to interact between sales and product and design and c-suite there's a lot of transparency across the org i think people even see that in the interview process when they come in.

1:13:02So I'll give a couple, I'll give a couple more because I'll let me in chat. It's like a shout out to Joe, Joe holiday, who's our head of mission. So he leads the mission operations team. Joe's phenomenal. Shout out Joe. Yeah, Joe, you know, is it was a Palantir for a little over seven years and sort of led a lot of the commercial pilot efforts there. And so if there's like a place, there's a dojo that you would have wanted to train in, in terms of like big, high stakes, complex deployments of software. Joe has done that at scale. And so I was able to, it's just a symphony watching him like work with our customers and get them value in nominal, like rapidly.

1:13:44And then Jen, shout out Jen. We have a very often sometimes when I contextualize the federal side of the business and like the federal footprint people are like wait where's the other 10 people um just given the results i think we've been able to deliver with a very very small team jen is like almost a one woman show there um and she's been amazing uh in in leading a lot of the the dod efforts um so now we we have an amazing uh amazing team and we're looking to to really double down um as we grow love the shout outs very necessary if you i i could just shout out the whole company it'd be but um people would be like what?

1:14:25We'll keep it. We'll keep it at that. We'll keep it brief. But all of these team members have helped you build an incredible product scale and then also raise over$100 million in total from some of the biggest names in Silicon Valley. Sequoia, Lightspeed, General Catalyst, Founders Fund, the list goes on. What have you learned from your fundraisers to date? And how do you think about milestones going forward? Yeah, I think, um, no, it's a really good, uh, it's a really good question. Obviously this, we, we started at the beginning, like this fundraise happened quickly, you know, 10 days kind of from start to finish.

1:15:06I think one thing that like a new learning, um, going through that process was we, we have a pretty killer, uh, you know, team and we, we were putting together these like very comprehensive response memos. you get a lot of rfis you get a lot of like questions you can tell our investors like really serious and digging in and we almost it almost became like a game um we're very serious about it but like we would put together 10 page responses overnight um just that's amazing yeah and it's like it should i mean i think it should i think if you're an investor and i've been an investor on the luck side like it shows that that person's serious um and i think being able to have this like trove of information that you can sort of expose to people I think was was one we were you know to be always careful like customers our customers are busy um and so you want to like protect ruthlessly protect their time but I think the willingness we showed also to get our customers on the phone like I would always just be like you should just talk to one of our customers like it will make sense to you um if you do that I think that um willingness sometimes when people are going through those processes like they're kind of like don't want people to talk to their customers.

1:16:15I'm sure I'd be like scratching your head and being like, I wonder why. We were just like, please let me get you on the phone with our customers because it will all make sense after that. I think it's a big one. And then, yeah, maybe we're just more general. The end of thoughts here is you want to give, they want to give investors data points that they can make sense of. Like they're going to be investing, you know, this is a Sequoia led this round. It's the growth team. They're investing in growth, right? And so you need to give them some data points and they will infer is the slope here is the slope here.

1:16:51And so just making sure that you're having those conversations before it comes prime time for a financing so that they know like, oh, wow, this team says they're going to do something and they will. I love the base episode. So I'm going to make a reference to violently execute. I think that was it. Yeah. Can this seem violently execute and just like go do the things they're going to say they could do. so yeah i think those would be the three three takeaways violent execution violent execution setting the gold standard yeah i love it that's great um so just a couple questions left um first i want to know where do you see nominal in the next five years what's your future outlook yeah i mean right right now i think with this this new you know pot of capital we're focused on you know a handful of things immediately i think kind of four of those to talk through but one of of those is like really expanding more and more into these larger enterprises.

1:17:47And to do that at scale requires a large team. It requires a lot of focus. And so we have a lot of really good traction with some more of these legacy prime providers, and we're going to move really fast to scale there. I love this headline, you know, and I worked at Applied Intuition. They had this incredible slogan that I think it was just, you know, 18 of the top 20 automotive oems in their case you know trusted applied intuition used applied intuition like crazy stat they literally had 18 in the top 20. and so i want like where we want for nominal is like i want that to be the case in certainly the big five you know primes but the same 18 of the top 20 i want 20 of the top 20 but you get the point um to be loving nominal and to be using it and to be sort of ubiquitous with what it is to to test in a modern manner and field hardware in a modern manner.

1:18:39So expanding to the enterprise. A thing that we're really excited about is talk a lot about, you know, the US lens here, but this is a global opportunity. Massively, we have international customers today, like we have successfully sort of done that motion. And so we're excited to double down and expand internationally. That looks like nominal offices overseas. And just saying yes to more and more customers, which is really exciting. A big thing, we talked about core and connect, but to win really big here, nominal is a multi-product business and multi-product strategy. We're keeping tight on a couple of those, but you'll be hearing more about them as the year unfolds.

1:19:27But delivering five, six, seven, eight products that all work together in this sort of hardware testing and operations domain, I think has been really exciting. And then I, you know, I give a shout out to Jen on our team, the one woman DOD show. Like we're going to be expanding and doubling down on the federal side. Like we've been really lean there. And frankly, the ROI has been really, really high, but we're bullish on a lot of those trends. And so we're going to be staffing up, opening a formal DC office later this year and really just building out, you know, that team. That is certainly what we're thinking about in the next, you know, 12, 15, 18 months.

1:20:06and then i think where yeah where nominal goes on a five you know the five year time horizon i think is just becoming uh i always say like the north star for us is when a hardware engineer generally speaking someone says like what is what is it that you do like what is your job um their friend might visit them and be like what do you do all day um they would open up nominal um and that's like the thing that's the vessel that they would use to explain like oh this is like what it is to be a hardware engineer like I have to fuse all these data sources from my machine I have to you know visualize them and like tons of it's automated but there's some things I still am an expert in so I need to like double check or explore this data and then I like build these these rules and validation logic that governs my hardware system when I deploy like that's that's the north star for us um and so we will just be building towards that vision well to close out one final question i want to bring you back to your roots and this you both can answer this but this is a this is a very hard question all right i'm setting the stage okay you're in the mediterranean there's a severe storm you're trapped on a very large yacht okay maybe the rogue wave comes by what do you do do you take a helicopter or do you take a submarine when you said the mediterranean peace site now i have this like is it wolf of wall street is that that when i'm thinking of like wolf of wall street i feel like you like got into um would i take a helicopter or a submarine uh i've been i say i've been on a submarine a lot so my inclination would be to take a helicopter but you said it's stormy and so submarine fun fact included uh when you're on a submarine very deep underwater you generally have no idea that it is stormy uh and so it's very calm with the exception this is a double fun fact um because i've done this and submarine people have as well but uh if you are in a submarine underneath like a hurricane uh you can actually feel the swells down to like hundreds of feet deep have you experienced that yeah yeah normally when there's like a hurricane i was stationed in norfolk so it was on the east coast like hurricanes come and they like sound the alarm bell and like the safest place for a submarine to be is not in the pier it's out in the ocean um and so you like scramble and get ready and you go out and you go deep um but just showing how powerful hurricanes are you can you're still jostling around to like 500 500 feet underneath whatever so um it's pretty wild but that's crazy that's a long-winded answer but um i think i would i think i would take a submarine but i want to hear what steven has to say about this i think i'm switching my answer because originally i was team submarine but i'm considering that first i have to get onto this rescue vessel and the odds i've never been described as sure-footed like the odds that i will make it onto the surface of the submarine and not just like slip off and disappear into the deep yeah yeah i'll be gone like i'll be gone in an instant and so i'm gonna take the easier you know hop into the basket of the helicopter yep and then i'll probably go down 30 minutes later in a blaze of glory it's very morbid answers we have i would hope not yeah because you're getting rescued yeah wait okay so going back to your job as a submarine officer could you just explain that a little bit further sure you skipped a lot of details yeah we skipped a lot of details um yeah i first part of my career as a submarine officer so what does that mean um Um, was there related to major jobs and almost all submarine officers start as an engineering officer.

1:23:54So the day to day is operating the nuclear reactor provides propulsion for the submarine. Um, when I say operating as an officer, you're not actually touching the things as much as you are, um, comprehensively sort of managing the state of the reactor plant and giving orders and receiving information so telling um others how to how to operate um that's what i did uh in the sort of engineering side for for a while and then the other job i had is i was our kind of assistant weapons officer so after you've done the engineering for a bit you go forward is what it's called but it's literally like the forward half of the submarine the back half's engineering the forward half is operations and weapons and tactics uh and then i was a weapons officer so um torpedoes mines uh tomahawk missiles etc but just having visibility over all of that if we needed to use them thankfully we my submarine did not um did not ever use them but that's a lot of the job um submarines do a lot of what's called isr intelligence surveillance reconnaissance so the mission set is mostly being places where others don't know you're there because you're very quiet and very sneaky and collecting useful pieces of information that others don't think that you're collecting um that the in this case the u.s can use to to get an advantage so many questions how long have you been under the water for i always i say this and then people are like what um i've been underwater 484 days not consecutively which most people don't see you did it too um It's still impressive.

1:25:34No, no, no, it is. Yeah, thank goodness not consecutive. That's my total, like, underway time. But longest consecutively was 84 days, which is still a long time. Okay, that's very long. Yeah, yeah. And most people don't realize, like, you generally don't communicate when you're under. So, you know, you learn things like if you don't get an internet sync with Spotify after 30 days, it just all of the songs that you thought you downloaded are not actually downloaded they go away like you learn these little little tips and tricks that um so it becomes very sad after a while when you have no internet connectivity and no you know comms with the outside world but gotta bring a cassette tape yeah gotta go did you see any monsters did you see any large squids uh i did not see any large squids but you actually do hear a lot of things so you hear whales um you hear dolphins you hear shrimp like you submarines all it's all sonar based and so you do hear a lot of sea creatures um and normally it's like fun because you're like oh it's like it's a whale or it's whatever but it can get they're very loud and you're you might be trying to listen to like another submarine or a ship or something that you need to and you're kind of like ufos yeah um you like pacific rim i'm thinking of pacific rim um and uh you might be trying to listen to stuff that's productive and then you're like these animals are like so loud um just can't hear what you mean quiet down yeah yeah exactly yeah okay well i've i've taken up a lot of your time and i very much appreciate it congratulations on your 75 million dollar series b massive and thank you so much thank you molly it was awesome to be here we loved it thanks molly this was great you're welcome thank you hey it's molly if you enjoy our interviews check out our newsletter sorcery.vc where we deliver a once a week top deals and tech headlines email and also go deeper on our podcast interviews.

1:27:33Subscribe to Sorcery today. And don't forget to subscribe to the podcast on YouTube, Spotify, Apple, or wherever you listen. Link in description to sign up.