AI's Trust Architects: The Patronus Approach to LLM-Based Applications and Enterprise Adoption with Anand Kannappan & Rebecca Qian
37:32 Share

The genre field and is the brand of applications that we could deploy AI into that have not been unlocked yet to stay negative. One out here from .

notable capable. This is found a real talk where we get real about the chAllenges that founders and start up executives face and how we've grown from tough c experiences. Found a real talk is part of the notable capital family of podcasts.

And you can access all our pod cats at notable cap dot com. I'm your host, glen salomon, without further a do here. Today's episode.

Dani are eaten tic to have joining us on founder real talk today. The two founders of patrol onus A I on in come upon and rebeca chen notable LED patronesses seventeen million dollars a series iron recently, and the company has been on an absolute tair on a rebeca have been up to some really interesting, an exciting things at the company. And we're gona dig into this past lots of other topics today on an Rebecca, welcome to found a real talk.

Thanks so much. Quite endure for having us for super, super excited to be on this podcast and really dive in to petra and and everything that were extremely excited. T but we thought .

we start with a little bit about you and your backgrounds and on and i'll start with you. Petronas isn't your first four ray into the world of ai. Tell us a little bit about how you got into this field, how you met rebeca and what let you guys to found the onus in general.

I've been my entire life really, really excited about all things related to the machine learning. But in particularly, especially the last several years, a lot of what I spent time thinking about is an an interpretation service and explain ability. And so in the past, I had worked as a scientist on the ocular team, at which today is the metal ground lobbs work.

And I do out a lot of the early ml foundations around things like causal and friends and events, experimental. And a lot of what I thought about is how do we make models a lot more usable, especially in a large organization setting, especially at maturity, loves. The organization grew from two thousand and twenty thousand people over few years.

And so sign up growth and seeing how lots of different kinds of partner teams were able to really scale what they were doing and really think about the ways to have to use machine learning was an incredible point that was really involved with and both of rock, and I would like to meet up, but you actually do each other from even before that, back in an underground. And so I remember voter reback and I wear an machine learning class together. And we spend a lot of time thinking about, in particular, not only where I was headed, but how we believe people on the world would use air responsibly. And that's really big power permission today and are super excited to be able to to bring that vision forward through the channels.

great. How about you, rebeca? Was there a moment that IT made to you to start a company with on and and tell a little bit about that?

absolutely. So I think there's two separate questions. That one is the moment we realized we need to start petronas and the question of when I realized we need to start a company unit.

So the the second question came away earlier, like unit had mentioned when we actually overlap when we studied C S. There to chicago. And we're working on to start up even even back then.

Actually my first impression of online was he was running a quite tch fan between a machine learning courses that was actually backed by my cuban and and so we we were ready working on machine learning problems back then. I said I knew I wanted work with on end. And we've at this point known each other throughout, you know, meta universe chicago and have worked together on on portraits for the pass you to have.

So the question of whether, you know, we we decided to start by, and when we realized that was the case was when charge of two was released. So gone you can remember, you know, back in late twenty twenty two, try to be tevas released. There was a lot of enterprise interests that at the same time, a lot of companies are banning IT.

And so at that time, like an invention, we were both had met. So while he was that made a reality lives, I was that a research of an on define, where I actually drew a response, A I, and develop this new pillar that back then you IT problems like close nations register, known in a research, studying, but of course, now become really commonplace terms. And so we just knew it's gonna be a problem. And that's, of course, i've been added for just the discourse that we see, you know the the headlines around a canada and shabby bots and just the are the reputational gress that mighty curve from lack of evaluation. So it's it's very clear that this was the number one blocker now to enter Price adoption.

Very cool.

Maybe give us kind of the quick background and what patro ony does. What's the the short elevator pitch?

Yeah I partners. We help companies scalability and reliably catch mistakes with alums and alone systems, including things like pollution instance and unexpected behavior and and lots of different kinds of unsafe outputs.

Um and like a reta said, enterprises have been extremely excited about the prospect of journey I especially over the last couple of years but there are also equally concerned about all kinds of potential failures and the petros. We have a unique air research first approach to how we solve that problem, where we do things like train evaluation models, deep limon techniques and new centralized benchMarks. And then we apply all that research into our product to ultimately drive real customer value. And so what enterprises and and all our customers have an excited about is the way that we approached the problem space and the fact that the product itself has yelled very impressed results that has now caught several thousands of polls, nations and other kinds of failures for for four, five companies as well as for leading air companies around the world.

So rebeca may be altering to you um you guys as you looted to earlier this is A I is not new to you um in fact IT sounds like the two of you have been even since the university days focused on machine learning and you know all all the promise that this field can bring when you when you saw like ChatGPT come out in G. A, I start to take hold what's been like some of the surprises for you and helps your perspective changed on where this this home, you area of technology and market can go.

So I I think I could go the rest the hour just talking about this alone. So I think the the biggest surprise coming from the researcher who worked on a safety and security issues to a practitioner in helping some of the world's largest enterprises with these same problems, as so in a research setting, so as some of the work, that is that a trained the first large language model fair with the with the finer subjective.

And we are very focused on demographic fairness, making sure AI doesn't discriminate against people based on their gender protective categories. That and those are very important problems and making sure models aren't not putting toxic outputs. And you know that, that was primarily what the focus on the events were before.

But then since working with enterprises and realized it's actually a much bigger problem even that an onour initially anticipated. And so some examples are brand alignment, like making sure the output doesn't reference a competence of dozens of violence in company policy and and that's different between each company. Another is the focus on your capabilities is like close nation.

So we, for example, put out finance spend, which the first large skill benchmark for financial analyst pe queries and and IT was actually very surprising to that a data said like that had not existed before. And so we realize, you know models were being assessed on like A A T benchMarks in middle math questions, and then they were being deployed in the real world. I am very complex, like financial analyst questions.

People see the promise you want to put put these these models to work in in applications and then realize, you know, oh, there are problems. There are potential risks. It's, it's it's truly remarkable how quickly this market has moved. sure. IT must be exhilarating for you guys because, you know, there is such a big need for what proton's is doing.

Yeah absolutely. Just just the the you know discourse around valuation and the tools that we have access to were also we saw that he was lacking. And so we want to provide more of that to to companies and preventing these kind .

of mistakes. I think folks have seen a lot of the impact of a lack of events in hy profile stories like the chevy chap, hot and others. But i'm not sure, you know, folks in our audience actually understand how evil s actually work. So if you walk us through the petroleum product, like what how folks using IT and what through errors and mistakes, can you help produce so broadly?

Ly speaking, the OCR onic platform focuses on two stages of evaluation, and one is predestined yet and one is post appointment. So so really you should start thinking about evil. And i'm working with us like the moment you want to develop in AI application.

IT could be a system that could be a chatbot. IT could be internal. And and so when you're doing pretty ploy ment testing, we have what we call valuation runs. So you can run, you know a batch of evaluations and that's basically testing your A I application around A A range of you know what we call valuation criterial.

So that could be like class nations like we mentioned, that could be toxic based IT could be, for example, in a rag system like you might want to test different aspects of bly context, relevance or making sure you achieving the information and practice actually really, really complex. And and actually, you know our rack evaluations perform twenty percent Better than industry and academic alternatives. So it's it's really difficult to get right.

And we're seeing that a assistance more complex. There is like many different touch points on the second part is, you know okay, you've deployed this thing into production. How do I make sure it's onions giving embarrassing responses to use? There's I am seeing embarrassing this is embarrassing when you're shoppy bott says that you can sell a car for one dollar and ford is Better rates.

So you know, we can catch those kinds of mistakes in real time. And as I sometimes we tell people in our goals to catch our unfAiling is mistakes at scale because we have this real time modern A P. I.

And so we are we are unique and that we are A P I. first. And IT is important to have commercially available variation API because there is no other way to be able to catch these kinds of mistakes that scale .

IT sounds like there's there's different ways to deploy patro onus on. And maybe you could talk a little bit about some of the customers. How are they using you guys and what are some of the use cases that you see people wanting to, to put you, deploy you into?

Yeah, yeah. I say that especially we talked a lot of developers over the last several months, one of the biggest questions we get is okay. We we, of course, understand that this is a really big problem and we're experience the problem and it's a really, really painful problem.

But why exactly is measuring alan performance so difficult? And what makes IT so different from pretis al mal predictive about, for example, and discount what we say pretty to my mall, you tend to get direct results, wherewith alarms. Because by nature of these models are generated, there is such a wide space of behavior that makes IT difficile to not just achieve testing coverage, but even just the final testing coverage even means. And so then what we've seen happen is a lot of companies have tried to solve the problem by spending a lot of time and money on manual evaluation methods where they have internal current teams, external consultants, even expensive engineering time allocated towards manually equating test cases and manually creating outfits and spatch its. And there was just something that we've seen companies continue to do just because they lacked confidence and automated approaches that just don't work well and are incredibly inconsistent and reliable.

And so the biggest thing that we do, a patronus, is bringing an extremely high quality, extremely highly reliable solution that is able to actually catch these mistakes in an in a very scalable way in both an offline setting and an online setting like a recommended and so today, our customers span various kinds of vertical, including uh, automotive, education, health care, of nancie services and and across various inds of types of things. Our focus always been to make sure that customers are able to feel a lot more confident and even compliant with the kinds of things that they ultimately they care about before they roll out a product, but also active role out product as well. In terms of our market focus, we specifically focus on companies and slightly more regulated industries because these are the kinds of companies that are trying to use A I A mission critical scenarios and therefore, the margin of air has to be extremely well.

And so so when enterprise is use us, those the kinds of things that they care about when they get started. But of course, from a part perspective, they care about a lot of different types of things, not just a quality like a mention earlier, but also flexibility and the response of the solution itself and some body to integrated txt anywhere across the stack. And I like a recognition, ed.

The fact that we have an A P, I for solution is really, really worth noting. There's because when you look up evaluation A P I on google, you you actually want to find anything. You might find some open source packages, but you don't find a company that actually offers something that is incredible, accessible to the form of an API. Those are some of the the ways enterprises have been using out some what you really excited about with our product.

It's awesome. And I think here you talk through that just reflects kind of how media this problem is, as when you guys came together to first found the company, enterprise adoption valves was still in its infancy. And here is rebec, especially given your research background, what gave you the conviction that this was gonna be a lasting problem? Knows the right place to focus the next no ten or so years of your .

career yeah yeah dearly so yeah likely said I wasn't fancy, I would say just as a whole. The journey field and the range of applications that we could deploy AI into that have not been unlocked yet is still we're still very much in early stages like on I always say it's stay negative one out here. So it's yeah so I say very much still the case today, twenty twenty four. And and you know in terms of you wanting to focus the next section of our careers to this problem, I I think really comes from, you know, even just in the beginning, like honey and I talking about how even back at chicago, we knew that A I was going to be transformative and we wanted to really dedicate our careers to A I and and from that, you know, realizing like what are the problems blocking A I to enterprise adoption. And this was clearly the number one blocker because he trust is really the the center of, you know, every everything that we day negative .

one is kind of a interesting way you can think about this market. And if you're right, if it's really day negative one, then, then it's just incredible because there's already so much activity going on. But I think on you know, we were curious to get your take. You're talking to tons of fortune five hundred global two thousand companies about petrol onus. We know because we've introduce you to a lot of them.

And IT seems like every single company we've introduced to is excited to talk to you, which tells us that there you know there somewhere between almost every company out there across industry seems to be somewhere in between experimenting with and really thinking about putting in a production L M based applications give us, you know, you have that catbird seat. So IT be curious to to get your perspective on IT. Do do you think that's really where we are? You know how far we weigh from seeing lots and lots of meaningful applications in production from large companies.

And you know what what what do you think the the blockers are? And you know what is the future? What is the look like when those blockers are deviated?

Yeah definitely. I say what's accurate, surprising and unique about this market at this point in time is the fact that large companies in the world are moving faster than ever. I remember around when we were getting started a year ago, I spoke to C I, O of a large minister company and he told me that they haven't book faster or on anything else in their entire histories since the eighteen hundreds.

And so seeing that kind of momentum was incredibly exciting. And and a question that we sometimes get is, you know why we currently in new york, and I say the biggest reason for that is just because we noticed this really amazing market opportunity, and we wanted to be as close to the market as possible. And given that we're focused on companies and actually more regulate industries, especially traditional buyers, we wanted to make sure that we were as close to that geographically.

And that's, you know one of the the ways that we've been able to continue develop our product over time. Uh, there are a lot of things that we've learned in terms of how enterprises have have, have tried to approach from space and voters with the blockers around a adoption. And I said there's a few different kinds of things that come up really often.

One is, of course, the pollution problem. And what's interesting is that love. People might agree that the definition of A I insecurity has been changing quite a bit where it's no longer just about the party actors and everything threats, but it's also about accuracy and reliability.

And that's something that has come up time and time again. And solution that we've developed in terms of our allusions section of voters we've able to show significantly outperforms all alternatives, even using tupi for as a judge by other twenty percent. And we're really excited to people to to continue continue to innovate all those kinds capabilities and even make those kinds of captain ties more accessible over time.

In addition to that, we're seeing a large genetic prises care about very enterprise specific capabilities when IT comes to evaluation. So that includes broadly various things related to brand alignment. So that includes things like the tone, voice of your chatbot or style conscious ness, bias, company policies, regulations policies.

And those are a lot of the kinds of things that, that larger companies tend to care about a lot more in addition to, of course, the the kind of reputation risk that they take on in the scenarios where the air products that they roll out ultimately have unsafe file puts across the board. I don't think that they are finding chAllenging is in a market like this, especially a new category. There is a lot of noise that's happening, and it's really confusing to figure out what to really do.

And that goes across the stack. And and if we will say, you know, start with the olympic side of things. A lot of enterprises are confused about how do you make decisions around their their album architecture stack, especially in a world where there are over a million models on hugger face.

Now all of the commercial companies like OpenAIr a nd a nthropic a re u pdating m odels e very t wo w eeks. And so I just become a really increasingly complex and environment to navigate. And so what companies have been asking for is an unbiased, independent company, almost a trusted expert, third party, I can actually help them navigate these kinds of of chAllenges and an extremely fast market.

And that's why there's article I came patterns around the end of last year. That called pattern is the moods of AI, where we are that unbias an independent company. And so not only from a product perspective, even from a company perspective, there's the kinds of ways that we ve been able to partner with large companies to be able to help them solve some of the most charging problems with alarms and .

the most powerful on the foot p side. We're seeing l ams, both hours of the largest enterprise in the world and also enable this massive wave of india velocity ment. I'm curious, as you think about the evolution, patronise, how do you think about supporting some of these all our customers or individual developers who are just getting started with elms but looking to achieve the same and reliability that enterprises are concerned about, concerned about?

Yeah, I said that the change were extremely excited to you to make our product a lot more accessible, especially over time. And what that actually means is we want to make the time to value through a product as as low as possible. And so we think that we developed a lot of really exciting ai driven features that, that ultimately make an evaluation security as easy as possible to do.

Um we talk a little bit earlier about wine. A P I. First approach is important. But in addition to that, we want to continue to make the kinds of offerings that we bring to customers as as as as diverse as possible.

And so one aspect of what were planning to launch in the coming weeks and months is what we call a value families. And so that essentially refers to different tears of how you can interact with the various kinds of of evaluation models and techniques are offered by proton's. And so in the same way that you have alums that are of different sizes, like myself, small and very large and cohere, small and coherent, large in the same way, will have those kinds of things through petroleum.

And so are you using small location? A kind of approach might really well for real time use cases as just because IT tends to be a lot faster and a lot cheaper with some trade off quality. But on the flip side, if you want to use a large of average model for china and might be a little bit more expensive and a little bit slower, but I will certainly have Better quality in terms of how good IT is that actually cashing mistakes and and those the concentrators that we expect. People to use large evolution models, especially and probably more offline use cases and to being able to to to offer to have different kinds of our offerings is, is one way that we want to continue to support any developers and make sure that folks get folks get value as as possible through the chance. Like he said.

the the fact that you offer the service via API makes IT, makes IT quite accessible, you know, even even for the end developer, which is exciting. I'm sure you guys feel like the whole market is your mister right now. Let me let me ask each of you.

I'll be back. I'll start with you. You know things are moving really fast and on and just talked about how models are changing almost by the week. And that's very confusing in chAllenging for customer ism gives you guys a role to play. You know there's still skepticism out there about the role of geni can play and how useful IT to really be as a technology. So we look we look forward like a year to um what are some of the signs that you'll be looking forward to say you like this is really working enterprises are achieving the kind of results that they wanted to achieve with with all this investment.

Yeah default. I I think right now, like you said, gone, everyone sees a lot of potential, but we're still in the early stages of large scale deployment. And and we think about large skills, societal deployment, we're talking you know, every single vertical.

And there's like some highest, some low risk like health care, medical, legal, finance, insurance a the possibilities are really unless and that's like both like consuming enterprise. So yeah, I death finites think we're not there yet. And in fact, you know, I mentioned this, like to predetermine tent and post appointment evaluation. What we would expect in the coming years as to see more and more companies go from the police deployment to post deployment when that's where like good time monitoring and having access to like in the ocean API is, is really critical to be able to test and prevent these kind of things that scale.

So for in terms of signs, you know that we are reaching that stage, uh, we would expect to see companies facing a different set issues like the discourse might shift from, you know, how to run these evaluations and both confidence and and you know shown totally that we can mediate these rests too. Like, how can I do this at scale? How can I prevent these things from going out to users? And actually, how can I make this fast as well? How can I do this and a low latency, cost effective way. And and we really haven't like a lot of the focus has been around accuracy and providing failures, but we're going to see you know more people ask about cost in in some cases, they might even be willing to trade off like a costly and see for certain levels of performance. So yeah, where we're really excited and where, of course, are very prepared for that have to happen.

I know. How about you any are there any clues that you drive from like your customers on on how they plan to deploy this type of technology? I'm curious like if the champion, uh, is typically a technical person that you're working with or more of a business person and what the interplay is between sort of technical and business .

oriented IT at your customers? Yeah, I say that typically the decision maker is a tech leader. Some kind could be a cio, C, T, O or some kind of management the year of the company.

But what's the unique as much about the current market is especially a cross enterprises. The teams are the organizations that are working on journal products today are sort of the shiny objects. And so everyone wants to have a piece of that everyone wants to be involved in and in shaping those new experiences.

And so we also have customers that are using our product that are product managers, designers, bd compliance, marketing, especially those are using our web platform in particular, just because you can run a lot of the same kinds of workforce, but pretty much without writing any code. And so we're using for the we're seeing, for example, customers that are setting up customer values or on patronus just by writing a sentenced to in english or actually any language for that matter. And they're able to define those policies in exactly the ways they want and their those off to developers implement them in the A P, I, for examples, are seeing those new kinds of workflows are really began.

And I say that to your earlier question. And in terms of the signs that, that we will continue to see what we tell enterprise leaders every day is you have to be as metric driven as possible from the moment you start and from the moment you think about what the product experience could even look like in the long term. And the reason for that is especially when you're doing with with models that are generated by nature, given that is a more difficult is easy to to forget that it's really important to do that and and see there's there are there ways to be able to to to measure that over time to understand if you are really moving the neural forward for your product and for your team.

Rebecca, this won be A H N A I focus podcasts in twenty twenty four, if we can ask you for some predictions. So, uh, would love to hear, what do you think the next big mind bling breakthrough will be from an A I lab?

Yeah absolutely is a near term and log term predictions. So yeah, I I think the in terms of the next big breakthrough, just overall IT would probably be a non transformers based architecture that is deploy that scale and and you know we've seem like some evidence of that would like to these models and thus been like talks have just like the next dominant architecture for some time, like we done so many optimization on transformers.

And that's really pretty much what like all the big a an airline providers are pouring money into right now. And so IT would be exciting to see some an alternative architecture compete. And there is definitely promise in in some of the like the sands, in some the you know architectures that people are expLoring with that like stanford, another labs.

I I think just in the implications of that. And also you know near term, like what we we would see is like basically a lot of a applications right now or so far that our successful have been primarily tax based. I think multi model has gotten more, more attention as a like image, video, speech at sea, but also in the future.

So I I actually was a robot researcher. And in the past I worked on in body agents and developing actually like in all you modules for robot assistance. So I think robotics are looking like down like the you know ten, twenty year horizon that the you know the big frontier for a of course, you need to have like proper data sing to test these things.

And so you know, what really motivates on day in and day out is this concept of skeletal overset, which is okay. We're onna. Enter this world where very soon A I is going.

It's already outperforming humans in any data date task. And it's these are super intelligent entities. How do you continue to supervise entities are more intelligent than you. And that's a constant scalable oversight, which is that humans are going to be acting as over series in a world where .

A I evaluate I okay, so so dan, the the future has been depicted for us. We're onna have AI robots and we're going to need scalable oversight. So thank you on an an rebeca for thinking about this problem before it's too late. We're gonna put you guys on the hot seat in end with our speed round. And then i'll start with you just say the first thing that comes to mind what what's the biggest misconception that you had when when you started patronus that you no longer have something you've learned about maybe running a company or starting a company.

I say that what's unique about the market that were in and the timing is that enterprises don't typically work. What start of they fear about their economic stability and and all kinds of things that might happen, the companies, but in the given what's happening today, they have to work with startups because starts by default, bulgars move faster.

And and along with that, the this isn't enough time or maybe even a focus for the company to to grow the AI expertise inside the company. And so they want to make sure that they they pick the right kind of long term trusted partner that they can work with. That's exactly where we will be coming. So I say that, that was probably the biggest thing that I deeply under reserved.

great. Well, certainly ms, like you guys have hit the nail on the head with respective timing. So Rebecca, how about you? What what advice would you give to founders who are thinking about starting something in in the A I. Field right now?

I would tell founders to keep an open mind and stay flexible, and they did not have any, priya, because, you know, what we're seeing especially now is that the market is shifting and moving so quickly. Never seen so many foundation models released in the past six months. There is a period when I felt like week to week.

New models are being released and continuing to push the frontier. So in research we always talk about, you know, state of the art. And like, what is state of the art today is not was not stood or what was still to a year go was not still the are today anymore.

And so you can have to be innovating. And on a nine or our whole team, we actually we love that because we love being environment where the standard are constant being risen, and we're always raising a standard for ourselves. Just IT really keeps everyone other times, as I until founders, is just to to be open minded, flexible and to keep up.

You have be ready to stay in your toes. okay. Last one and i'll ask you to both your rebeca start with you. Do you have A A frequent, you know, in your personal life or in business like A A L M U case that you you enjoy oh well.

I I can't share ilm use cases in our business. But what I I think first stall, no, we we use a lot, of course, you know, in in all aspects of development, our platform and but in my person al life, probably yes, and I am a youth case for me is is actually around like creative applications.

So just anything because I I think and that's actually started, you know, when was research just like I think what that language moes can really help humans with is like generating ideas. And so you like IT gave me a name for for something or like you give me some ideas for like a birthday gift, something I cook in. And so just around and I like give me like list of. You know, different variations, names and recipe and things like that.

Great about you. How about you? Anything, anything, especially you you've used l ms.

For yeah, I promise you we did not coordinate this. But I was about to the exact same thing around idea generation. And in particular, i'd say, uh, in particular, what has been really helpful for me is actually generating ideas around the code has been has been incredible since the beginning.

I was actually one of the early beta users of get up copilot and twenty twenty one back when I was offered for free. And those one I remember not just inside your idea, but even as using charge. Pity is a general called recommendations that was a really big solar. And for everything that that I was building. And I in addition to that, maybe a second one that has been more helpful for me in recent times as summaries of a books and podcast, maybe like this one that that has been incredible just because sometimes I just want key takeaway really quickly and I want to just get maybe the most interesting of the most surprising insights as fast as possible. And and that's been a great way for me to just in quality to stay on my toes on in terms of everything that I should be thinking about and and the .

kinds of ways that i'm learning and growing as well. That's fantastic. Well, on a rebeca, I you know speaking up a half of them and everybody at notable capital myself, we are so, so excited to be working with you and the entire portion ious team.

We really appreciate you guys coming on today and sharing your thoughts and perspectives. And IT gets us even more excited about the future. We we look forward to big things from patronus as we know you do, and we're really excited to see where this goes. So thanks so much.

Thanks much for having a this incredible and it's been amazing partners with you all were super, super excited for twenty twenty four beyond.

Thanks so much for listening to another episode of found a real talk we really hope you enjoyed IT. If you like what you heard, please rate and review us on your podcast APP to help others find this podcast. If you have any questions you like to ask our guests or founders you'd like to hear on this podcast, feel free to email us at F, R, T.

At notable cap c, notable capitals, a global venture capital firm based in the U. S. Focused on early to grow stage companies in business and consumer applications and cloud infrastructure, which includes A I cybersecurity data develops and more. The firm has officers in silicon valley, sanford, go and new york invest primarily in the us, israel, europe and later america. Noticable capital enterprise portfolio des companies such as bedside block, disco de jordi, hosea, mcara ion mazoe networks, orka security, slack sinai for cell and desk and many more learn more at notable cap dot com or follow us on x and linked in at notable capital.

This episode was produced by sasha productions, music produced by guas kobo. The opinions expressed in this podcast are those of the individuals only and do not reflect the views of their respective employers or pilots. The information provided is for informational purposes only and should not be considered or deemed to be as investment advice in anyway, thanks again for tuning in.