Episode Transcript
[00:00:00] Speaker A: Jim, what a pleasure to have you today. I really appreciate you joining the show. Something incredible has happened over the last couple of years, and I have a confession. To start with, I've always said there's no such thing as AI. It's sophisticated statistics, it's nothing real. But I was in shock when Chat GPT came out, and that was the first time that I said something has changed. You're doing something incredibly interesting today around generative AIh. Tell us a little bit about what you're doing today.
[00:00:30] Speaker B: It's funny you say that, Ari. I came to much of the same shock and awe. So, by way of background, I've been a serial entrepreneur for most of my life. I've started and built over a dozen software and information technology companies over that timeframe, and I've seen a lot of great tech and have had some great successes and some not so good successes.
So we started a business about ten years ago called Fintech Studios. At its core, it's a platform company that's focused specifically on using AI, machine learning, NLP, all of the new cloud technologies, and massive amounts of unstructured data. Our view is that a lot of the world's problems are going to be solved with better organized, better structured data, but on a global basis. Part of our strategy has been to ingest and tag from millions of sources, sort of like a Google on steroids, but for the professional market, but then to use machine learning, AI, natural language processing, to take all of that unstructured data, to tag it, to structure it, to organize it, to put it into structured databases, to run analytics against that data, and to be able to use it to search and generate insights and generate intelligence. So we've been a huge fan of using the most advanced stack. In fact, all of my prior companies like Multex, you know, 25 years ago, was using the most advanced stuff when the web first came out and people were starting to do some really crazy things. But what was interesting is most of our effort around AI machine learning, NLP was all in the back end of our platform, all in the quality ingestion pipeline process. This thing that at Multex I had 600 people manually tagging and indexed brokerage research and fundamental data and so forth in much smaller scale, I mean, hundreds of sources.
The whole business before we took it public, only had millions of reports in it, and that was amazing. Today we generate millions a day from millions of sources. But the other trick is languages. We've really figured out how to take languages in any language, German, French, Italian, Chinese, Korean, you know, you name it and not only find it, but translate it into any other language and generate insight out of it.
When chat GPT came out, what was fascinating is we had already built some really solid b, two b market intelligence and regulatory intelligence platforms. We had a large and growing client base that today has over 800,000 users and all of them licensed b, two b users with big clients, like a Citibank or a JP Morgan, or an S and P global or an accenture. But the products have just advanced so much and the capabilities with the advent. So, fortunately, because we had this massive data moat of billions of documents over the last ten years and trillions of data points, part of the problem, as you probably know, with large language models, are hallucinations and relying on these models as sources of truth. They are not databases. People's fallacy is, if I ask it to give me factually correct data, either it's old information, because they don't update in real time, they update annually or periodically, but b, they don't structure data in ways that you need for the sort of stuff that we do. So when chat GPT came out, it was revolutionary in my mind, because all of a sudden now you could start to, in both a structured way, but a conversational way, sift through and find sort of the needle in the haystack, this insight around companies you're tracking, or people you worry about, or sectors or industries you care about, or the portfolio that you're managing, or ESG or crypto, or anything that we track, because the world is changing so rapidly, no human can keep up with it all. And so we've launched a product called Power Intel AI. It really is. Think of if you're familiar with products, which I love like perplexity, and all of the chat GPT like equivalents from Google, like Gemini and Claude from Anthropic, they're all really cool, but they're not built for professionals, they're built really for consumers. And when you put it to the test, I think what you're going to find is there is increasingly this view that Genai particularly is moving even more aggressively because of the problems that you need to solve into vertical markets rather than just consumer markets, because that's frankly where most of the money is spent anyway in this. But yeah, it's revolutionary. I would put it up with the advent of PCs. When that first came out, it changed the way businesses operated. And when the web first came out, with Mark Andreessen and Netscape and the first browser, and then online services that spawned from that and e commerce services, and now everything is online and social media has exploded. And then the smartphone was sort of the number three in my list of revolutionary technologies that change the world. And I would put generative AI, even in its very early raw form, into that same category.
[00:05:46] Speaker A: I completely agree. I want to dive into who is actually using your customer. So it's basically a business intelligence platform. Tell us a little bit about the day to day of who the consumer is, who the user is, how they use it, and what is the problem that they're actually facing.
[00:06:05] Speaker B: Sure. So the products and services we offer cover two core areas of focus, market intelligence and regulatory intelligence. We deliver this via the web. We deliver via full stack applications. We deliver via integrated dashboards or widgets or APIs. So we can deliver it a variety of ways, but at the essence, people are struggling to sift through all the information to find meaningful information, both in the public web, but also within their own data stores. Finding contracts and the right policy information and figuring out what's material and what's important is increasingly difficult. So if you span across some of our clients. So, for example, Accenture is a very big client of ours and a partner of ours, and they use our products mainly for regulatory risk and control and compliance, not only for their own businesses internally. And they have a number of use cases, including anti money laundering and risk monitoring and compliance management on behalf of clients. But they're also selling it as part of a branded Accenture set of services wrapped around other technology and services they offer into major enterprises to solve the same problem that most corporations are struggling with. They can't keep up with all of the things that relate to their business, their products, their services, their competitors. And Google doesn't cut it. And perplexity is cool, but it's not built for professionals. And so there's this big gap. And what that means is people waste enormous amounts of time looking on Google for a list of links, looking at all of these things that may or may not be relevant, and they don't have specialized ways to organize it or specialized ways to analyze it. And so it's ultimately about productivity enhancement. It's ultimately about surfacing key market intelligence and regulatory intelligence in a timely, accurate way that in most cases today is done manually with teams of people, often expensive labor or disparate systems that try to piece together these capabilities.
So there's that. But then when you dig deeper, we have clients that are financial advisors that use to track portfolios of all of their clients or look at major trends in the market. We have users in investment banking and in sell side and buy side trading, and in institutional sales, and in venture capital firms and private equity firms. And we have large banks who deliver our intelligence to their retail investors through mobile devices and through their applications internally. So it ranges widely. But it comes back to who do you know that doesn't use Google every day? Because everybody is always searching for information and intelligence. And that's the other thing is we've cracked how to bring in millions of sources across languages that at Multex I only dreamed about 25 years ago, but the cloud tech stack, and now with generative AI, I can find the 20 documents that matter to you and anything you care about and in seconds summarize it using Genai and generate a digital newsletter or a set of highly targeted summarizations. And I could do it in any language, in any construct.
And that has enormous implications around how people author information. Now, in fact, many of our users use digital newsletters because they're so time compressed. Even with the best search tools, they would rather have a specialized channel be set up for their own, what they care about. So if you looked at yourself, Ari, what do you care about? And whatever that is, I could set up one or multiple channels that once you set it, you can forget it. It'll constantly update in real time. It'll notify you via alerts or via daily updated personalized newsletter. That's all done with generative AI because we synthesize and summarize it, and in 30 seconds you can sift through what might have taken you hours or days to do manually.
[00:10:04] Speaker A: So here's what I'm hearing. I want to kind of simplify it to bring all the audience with us. We have massive sources of data. The way we kind of went about it in the past is we put in a search query. We got thousands, hundreds of thousands of results, but really we only looked at the top three, four, five, if we're really being thorough, maybe 20.
And then we had to go into each website and kind of see, does this make sense to us, really? These people are trying to sell us on something and kind of find the nugget of truth. Hugely manual and intensive process. What we're shifting to today is, yeah, chat, GPT can answer stuff and information based on the information that it has, but it doesn't necessarily have the ability to go out and do research, and it can't necessarily use my private information that my organization owns. And even if there is a way to kind of upload a few documents. I'm kind of concerned about privacy, compliance, PII, stuff like that. So there's a need, there's a gap in the market for a professional tool to help us not just look at the first 20 results, but look at all of my data and find a very specific nugget inside of that data, and then put it through a very specialized process where I'm trying to get an output of a very specific need. The uses of these are not, you know, it's not for marketing purposes. These are, as I understand it, for very serious decision making purposes. We know that artificial intelligence suffers today from hallucinations. You mentioned that before, Jim. Have you put on top of these AI technologies mechanisms to prevent the hallucinations, or is it still the on is on the human being to kind of fact check everything that AI generates?
[00:11:48] Speaker B: Well, first of all, no matter how well you do this, there's always a degree of things you have to factor into the way you work with large language models. There's just no question about it. So for us, I think the important things are, number one, we don't rely on these large language models for data because we have a large data store that has billions of structured documents. We use what's called rag technology retrieval, augmented generation, which means when you make a query in our system, find me the latest news on Apple or ESG or my portfolio, or the regulatory filings about the latest proposed rules about criminal whatever. What ends up happening is we don't go to the models for the answer. We go to our databases. We use our advanced searching and algorithms to find the most relevant documents that answer the question. We then use our metadata that's been extracted that we generate when we index those upon sourcing, and we feed that plus a well defined prompt around that metadata to the large language models. And we tell it specifically to only use the data we give you to generate an answer. And so we know we have factually correct sources, we know we've tagged it extremely well. We've had years of users using this, and we have millions of queries every month of people banging away at our system. These are pros. Like you said, if the quality of the data sucked, we'd be in trouble. And our business keeps growing. Our user base, while not perfect, it's amazingly accurate, it's amazingly timely, and we've solved both the timeliness issue with large language models. There was a point not too long ago where even chat GPT was a year out of date from its last update. It's not very good for timely market intelligence about you or your peers, your competitors or whatever. Now it updates it every, I don't know, three to six months. But they don't have data structures, really. And when you look at things like perplexity, that's even worse because they rely on, as do others, Bing's search API to resolve a query. So they go up, they get the latest eight or ten docs that Bing search API gives them back and they use that to try to create the response, which is beautiful, I love it, phenomenally cool. But when I've done some side by side tests with them and others for business users, they're often completely inaccurate, partially correct or hallucinatory. And it's because they don't have data that's been well structured and organized and they're using basically just text. They're grabbing and hoping they can piece it together. And often if it doesn't have an answer because it doesn't have the right data, it makes it up.
[00:14:33] Speaker A: I mean, that goes back to the old saying, right? Garbage in, garbage out is simple as that.
[00:14:38] Speaker B: Oh, it's very true. It's very true. I mean, all of my businesses have been primarily in my career, information businesses that are high value, hard to access data.
People will spend a lot of money to get to the right nuggets at the right time. And sort of the way you think about it is there's always been the sort of the horizontal search engines like Google and Bing, and they're phenomenally successful and phenomenally useful, and I still use them every day. And then for millions of users in professional markets, there's professional search engines, there's FactSet, there's refinitiv, there's Bloomberg, there's S P, there's LexisNexis and law. There's all of these verticals that people spend large amounts of money. I mean, a Bloomberg terminal is $2,000 a month and yet they have 350,000 users who are willing to pay it. And that's true for a factsetter and s and p. So think of it as vertical search. Why are they in business and not Google? Because Google doesn't do what they do and they don't do what Google does. We try to really bring the two together and have a very specialized deep vertical, but using highly structured global data in multiple languages and then applying Genai at the core. And so we think that is a big differentiator and that solves a lot of the problems because our biggest problem in life is we don't have enough time, we don't have enough time to get to the answer. And for businesses, the wrong answer could put you out of business.
[00:16:06] Speaker A: I think it's incredibly interesting. There's a lot of fear, I believe, around AI that it's going to take the human being out of the equation. But what I personally believe, and I'm seeing happening, that really it's empowering individuals to be smarter and really define their strategies and tactics and then have a lot of manual data just be done for them. I would almost argue that the AI tools are really going to be putting people out of business only if they're not going to use them. So if you either stay behind and get left behind, or you kind of embrace and become more productive, and that really becomes your competitive advantage. I wanted to ask you about the future. How do you see industry, financial industries? Obviously, in your case, how do you see them changing with the introduction of generative AI?
[00:16:58] Speaker B: First of all, agree with your first comment. Maybe I'm always the glass half full or the optimist, but having been around long enough, I can see already what this is going to do. It's going to revolutionize every job in every industry, even in farming or manufacturing in places now those are slower to adopt. If you look at the adoption, it's mostly in tech companies, financial services companies, more advanced companies, but it's starting to accelerate. So I agree with you. First of all, you have to embrace the change. If you didn't embrace PCs when it came out, you wouldn't have been hired, you wouldn't have been promoted. Your business wouldn't have grown if you didn't embrace the Internet. I can remember when I launched Multex, which was one of the first web based delivery brokerage research in the world with Goldman Sachs and Morgan Stanley and Merrill Lynch Research online. That never was done before. And I had a lot of brokers saying I'll never put my research on the Internet, ever. It'll never work. And now all of them do. It is their core way of delivering it. So I think you have to embrace this technology. I think it will change your job. It will make you more productive. It's really, they're not going to take over your job, right? They're going to assist you. What I just did earlier, it would assist me in what I would have spent hours doing menial searching and trying to sift through 20 blue links on Google and reading every article and trying to figure out if it's even relevant. And that's after I sifted through all of the ads and sponsor links on Google.
So I think the future in financial services is full on adoption of AI in existing workflows so that you can start to add incremental value.
And so it's true, whether you're researching a company or tracking the portfolio for your clients, or doing regulatory horizon scanning on what affects your bank or I checking on m and A opportunities, or checking on what's going on in the private markets or whatever, it will just allow you to do your work better, faster, more effectively. Imagine if I could output the same thing in 14 languages at once and now reach a whole nother audience in a digital age without zero extra work. Right? So I think it will change. I think if you don't step up, you run the risk of being antiquated. Right? It's the old horse and buggy versus automobile story. At some point, if you didn't shift gears, you better find a new industry to work in, because horse and buggies were going to go away. And so jobs that are easily automated, with agents and workflows and the right human guidance and support, those people will need to elevate their skills and probably be the ones to create the prompts. I think, for example, prompt engineering, it's a phenomenally attractive area.
We put a lot of energy into it to get the prompts right, to get the outputs right, to get attribution right. It's not easy, and I don't think the average human is going to spend a lot of time writing a long prompt with a lot of stuff he doesn't quite understand. So I think it's going to be everywhere. It's already in your watches and in your phones, and you see it now in all of the new advances coming out by all of the major players. So I think it's a question of embracing it and realizing it's not going to take your job. It might make your job more valuable, it might allow you to do more, to reach more customers, to do it more productively.
There's even an article I read which I found fascinating, that with all of this technology, with three really smart people, you could build a three person unicorn.
A company worth a billion dollars.
Multex. I got to a billion five, but it had 600 people, and we had raised 50 million of private capital and another 50 plus million of public capital, and it took ten years. But 600 people. Only reason you can do that now is this technology.
I don't think you should be afraid of it. I think you have to embrace it, whatever your job is, because it will help you do your job. Better or find the next job if you lose your current job.
[00:21:00] Speaker A: It's so interesting. I mean, this has been since reversed almost immediately, but at some stage, Italy actually outlawed Aih. So they actually, I saw that and I was like, this is the stupidest decision ever. And they have since come back to it, and to their credit, very fast. But my question being is, what do you see the role of government? Should they just stay out of AI? Or is there a role for government to be involved here to empower it, to regulate it? What's the risks and rewards here?
[00:21:31] Speaker B: It sort of speaks on a broader topic, which is really the global political environment that we're faced with now. Because fortunately, unfortunately, AI in all forms, including Genai, including open source Gen AI, is now available not only to western economies, but also others that aren't friendly to our economy, whether it's China or Iran or Russia or North Korea and others. And so I think as you think about this going forward, governments are already involved. Like it or not, there is a role to play for governments, to provide guidance, to provide guidelines, to provide standards, to look out for the interests of consumers. I'm glad to see that some of this work is going on. I think in some places I sense that they're overstretching. And also I think they could be, if they're not careful, they could have a disincentive that might actually help other countries and other economic systems leapfrog us. If we restrict and constrain through over regulation, a burgeoning new birth of a major industry that will change the world. Right?
Certainly there's concerns about everything from copyright to privacy to hallucinations. But I will tell you, when the first PC came out, they sucked. They were terrible, but at the time they seem magical. But you look at them now and they're like, you know, black and white, no memory, no disk, slow, no online. And it's true with everything. The first Windows operating system was terrible and buggy, and it took years for apps to come and work, but you could see the promise of it. Same with the first browser. You could see the promise of it. This one. Governments need to be involved. I think it's important that they work with both major players in the industry. And my other concern is because the big are getting so much bigger, because the capital is now dimensionally so much greater. Like OpenAI just raised six and a half, $7 billion on $150 billion valuation is we got to make sure the little guy, the startup, the entrepreneur, the small business operator, still has a place in this market and can't be over regulated to death. Or the bar and the standards which they started with, I think five major foundational models. And then there's some that want to push it to anybody doing it, in which case, so I think you want to be careful with it. But I do think we need some level of government oversight. I think we need to coordinate with our allies across the globe around, because if you have different standards in every market, it's chaos for companies and that's more cost, more regulation, more oversight and less innovation for investing in the things that will actually help consumers.
[00:24:20] Speaker A: I think that's such an incredibly important point that you're making because we kind of think that, oh, you know, government, they add this regulation, oh, that's protecting us, that's going to help us. But in some cases that is true. But in other cases what that really does is protect the incumbents because the incumbents have that money, they have the ability to comply. So what you're really doing is you're preventing new entrants, you're preventing entrepreneurs, you're really preventing innovation. And by preventing innovation really you're preventing competitors and thus you're preventing bringing down prices and commoditizing services. So in many ways I think, and.
[00:24:59] Speaker B: Then they complain about these big oligarch, either monopoly structures or massive companies that they then want to break apart, that they effectively help create, help build by creating regulatory costs and efforts that are so oppressive that you need so much capital to comply or certain staff to deal with it, or the big players that have enormous capital and are throwing more at it every day. They're able to get past that piece and keep advancing, whereas smaller firms won't even get to the starting line.
[00:25:34] Speaker A: That's right. That's right. A great example of this is the compliance, which you're deeply knowledgeable about as well. Just to get one of these more, let's say standard compliance standards, SoC two, ISA 2701, even GDPR, you can spend minimum of six to nine months on getting that done and you're spending hundreds of thousands of dollars to do that, which I don't know many startups right to, you know, a guy and a girl and a dog in a garage that can do that. It truly is a barrier to entrance.
[00:26:08] Speaker B: One of the reasons, by the way, our regulatory risk business is growing is exactly for that reason. The number of laws and rules and proposals and amendments and all these things are exploding and every business, including the small ones, can't keep up with it. And so it is a big problem and if you can automate some of that, that helps. But I think you got to be very careful on both a local, by the way, this isn't just at a national level. States have their own laws and rules and guidelines. And you saw California was trying to enact some recent laws that the governor didn't approve, that there was a lot of resistance from technology companies as to the impact, negative impact that that would have had. And if California, which is usually a leader in new forms of consumer protections and ESG and climate related stuff, starts getting into this aspect, having been the birthplace of most of the AI and stuff, which also is sort of weird, where do you think that would go with other states? Other states would follow California, and all of a sudden you'd have a much different regulatory climate at the state level. And so now you'd have to.
[00:27:14] Speaker A: And we've seen that happen about state.
[00:27:16] Speaker B: By state differences, that how can a small firm, possibly, particularly an online firm that does business anywhere in the world, now have to figure out all the things that might be happening at the federal level, at the regional level, like EMEA or Europe as a group of countries, or now at even the state level or even at a local level.
[00:27:35] Speaker A: That's right. And we've seen that exact description play out with GDPR. GDPR started up in Europe, California adopted it, and then now we have eleven additional states which have adopted either the California standard or, or a version of it. So we're seeing that exactly play out. In creating these barriers to entry for the little people, is there any bright light on what might help the more entrepreneurial small companies that are trying to really bring, benefit and innovate? Do you see any bright lights coming to us in the future from that perspective? From an optimistic standpoint, I'm very optimistic.
[00:28:14] Speaker B: I think this country is. From day one, it's all been about entrepreneurs and innovation and startup and technology has often led the way. I think this country is the world leader in almost every technology.
I would hope, honestly, we bring manufacturing back to this country, and not only for job employment and all those things, but it builds infrastructure and skill sets that can be redeployed for a lot of these new age things.
I think the reason I brought up the three person unicorn, I think the right people with the right tools can build anything today. And because they move, the reason we are successful is we move faster than almost everybody. We're innovative constantly. When chat, GPT came out, we were on it the next day. When the 40 model came out, we had it in production. The next day, most of our clients didn't even know it existed, had no clue how to do it, and would be trying to do this months or years down the road. And the problem is things are moving too quickly. And that's the other reason, is while there's a lot of money out there, both in the venture community, right now, the markets are a little fickle, but that will change. I've been in these markets for 40 years, and there will be another day when, for the right reasons, more of the private capital flows into more and more startups. Plus, I think you're seeing this explosion of secondary markets. Right. The IPO markets are struggling, but there are private markets and secondary markets. The reason I bring all that up is you need capital formation, you need to raise capital. But with technology today, you don't need to raise ten or 20 million to start.
[00:29:46] Speaker A: That's right.
[00:29:46] Speaker B: Perplexity started with a few guys in the lab that had a lot of good ideas, and they built a prototype and within a year they raised $27 million. Right. So it's still doable. I think there's a lot of hope. I think the regulatory issue works itself out of, or like with me, you build a business around knowing that regulations will keep growing and expanding and trying to solve, because there's tons of problems just in that area that you could focus on. But I would also say we're in a global world that everybody will be doing this, Genai will be everywhere, and all you need to do is have a sliver. Think of it as a very thin slice of a vertical that you're just really good at, and you've built the right data and people and products, and it may not be the next facebook, but it could easily be a successful venture, a successful project. So I'm very optimistic. I think my biggest challenge is with the corporate structure, both in the United States and other sort of traditional industrial countries. I think the corporate structure is suffering. Right. The bureaucracy, the stagnation, the lack of. Now companies fire entire teams to meet their quarterly numbers. There's no lack of, there's a lack of trust and confidence. And so the best people really don't want to work in corporate structures of any consequence. Which further puts the opportunity into startups. Right. If they're firing people, you'll find often in recessions when people are letting people go, that's when many of the biggest startups were formed, because the best and brightest, when often started their own business, is often to compete with the companies that fired them.
[00:31:23] Speaker A: I would argue that to a certain degree recessions are a great time for companies to invest. The cost of labor goes down, you have basically the opportunity to grow and then kind of come out with a bang once the recession is over. So in some ways, it's even an opportunity if you adjust accordingly. Jim, I want to ask you a difficult question. What piece of advice would you give 20 year old Jim?
[00:31:48] Speaker B: What piece of advice would I give 20 year old Jim? So when I look back at 20 year old Jim, I was just in the process of starting my first software company as a senior at RPI, which is a technology and engineering school in Troy, New York.
And I had done some co op assignments because I was struggling to figure out what the gym of the future was going to be. And I originally went into school as a chemical engineer and did some co op assignments. And I realized I hated engineering, but I loved computers. So when I came back, I changed my major to technology management, computer science minor major. And so I guess the thing I said to myself was, you have nothing to lose by starting your own company at 20 because you can always go and get another job. And so if I look back at it now and say, what would I tell myself now that I didn't tell myself then? One would be, don't work 120 hours weeks. And I was, you know, I didn't, I wasn't married, no kids. You know, I had girlfriends, but I was working seven days a week, 1215 hours a day, to the point of obsession. And I would say, now that's maniac. There's a life balance that you have to think about.
I would have also said, focus. 20% of what you're doing is going to generate 80% of the results.
Now I'm very big into focus. Focus and everything else you can't deal with, even in scale, even as a big company, you need to focus because markets are moving fast, technology is moving fast, the culture, everything's moving quickly and with degrees of uncertainty and then be flexible.
I would say early on, you think you're smart, but you realize now you were pretty stupid, you were pretty naive, but you were aggressive, you were motivated, excited. I mean, I started at the time Bill Gates was starting Microsoft and so forth. So I was like, this is going to be the biggest, I mean, I really believe this was going to be. And it ended up being the biggest revolution known demand. But I also would say, raise more capital because as a 21 year old, I raised a million and a half dollars, which got the business off the ground, got us into some products and services, and revenues. But ultimately when we started competing, when software started being distributed more than just in brown paper bags and a little computer store, but with big ad campaigns and marketing without the capital. And so again, those would be some of the lessons and things I would say to the gym of 20.
[00:34:23] Speaker A: I appreciate that. With your permission, I'll add some of my own experience.
Also a serial entrepreneur. I'll say one thing, the companies that have failed, even in failure, I have learned ten x more than I would have learned in failing than I did in a year, let's say, of corporate America. So if you're a 20 year old youngster doing your own thing, you will learn so much that even if you fail and you decide to take a job a year or two later, you have so much experience that will be valuable for the corporate environment.
That will be, I would say, ten x 20 x more valuable than anything else. So I.
[00:35:08] Speaker B: It's funny you say that. It's funny you say that. That's exactly what I said to myself at the time. I got nothing to lose.
[00:35:14] Speaker A: That's right.
[00:35:15] Speaker B: I'm going to be more valuable if I gain all of this experience. And I will tell you those hundred hour weeks for three or four years. I gained 20 years of what I would have experienced in corporate America in three or four years. And I went on to start another one and another one and so forth. So yeah, I had nothing to lose, everything to gain. And that first company, mirror Images software, did not succeed after I left a year and a half later went out of business.
But the experience I gained, the lessons I learned was so much more valuable than if I taken a corporate job at IBM or GE or someplace working in some junior position. So you're absolutely right. So I see that as really a way to think about what are your risks versus your rewards. You got nothing to lose. Nobody's not going to hire you. If you came from a good school because you spent a year or two starting your business, they would think that's actually a good thing and probably hire you in a more advanced position.
[00:36:12] Speaker A: I completely agree. I would echo that. Jim, it was such a pleasure talking to you today. We covered a lot of interesting topics. I appreciate you coming on the show.
[00:36:21] Speaker B: It's my pleasure. Ari, thanks so much for taking the time.
