00:01:57 · Donald Muir
[00:01:57] S0: So dating before ARK in my prior life when I was working for corporations, I grew up in traditional private markets investing. I worked in middle market and late stage private equity buyout. And so the pain point that I am now solving with AI, I experienced firsthand in my prior life late at night and long weekends in the bullpen working for large private equity funds, doing commercial due diligence, modeling investment committee memo generation for large financial sponsors. All the workflows are completely offline. It's Microsoft Suite. It's Excel. It's Outlook. It's very inefficient. Fast forward, go to Stanford Business School, found my first company, ARK. As you know, maybe for everyone else, you can think of ARK as a digital Silicon Valley bank, but its roots were actually as lending business. Before building the banking business and the debt marketplace, I actually started out as a direct lender. So I raised hundreds of millions of dollars of debt alongside my venture rounds, And I lent money to other startups, and then I collected it. And so I also got to experience the pain points of direct lending. Went on to build out ARK. So turned the lending business into a debt marketplace with hundreds of private credit funds on the supply side of the marketplace. And then with my banking business, eventually, thousands of tech companies on the demand side. And I was helping the two sides meet each other and effectuate debt transactions. So the lenders would lend to my banking clients at ARK, and they continue to. Well, to make a long story short, with the advent of generative AI circa late twenty two, early twenty three, I saw an opportunity to automate the debt placement experience on my debt marketplace. And so we get loan applications from banking clients, and I built AI to take that data, standardize the very messy kind of Excel based dataset from lower middle market companies, and then create custom screening memos to surface those debt investment opportunities to the lenders. So building out that technology, we were really early just organically developing TAC to make our own debt marketplace scalable. And it did. And so we built TAC and allowed us to go from tens of millions of loan volume in ARK to tens of billions of loan volume, thanks to this AI native underwriting system.
00:04:10 · Donald Muir, S1
The moment for me was when the lender is particularly up market on the marketplace said to me, we wanna use your AI for not just the deals you're sending us in the marketplace, but for all of our deal flow. And in order to do that, I needed to start a new business. It was non core. Building out that product would have been very, you know, non core for ARK. And so I went back to the investors and and my team and my customers and saw overwhelming support to start what is now F2. And so the new company, F2, AI for private markets, has a bunch of former employees from ARK. Effectively, all of the investors from ARK, 50 of them participated in my first funding round at F2. And a number of the customers, the lenders I worked with that are now customers at F2. So lots of overlap on the other side of this, and the two business now have pure operational focus on serving their customers to the best of their ability. [00:05:03] S1: On the topic of customers, I know a lot of founders who work with ARK, and I've never heard anything but really, really good things. And I feel like that's very hard to do as a brand. You know, normally, it's like, it's mixed. The, you know, 10% will be hostile or say they had a bad experience. I've never had a bad experience. So think that says a lot. But also it makes me wonder, was that a hard decision? Like, that can't have been easy to make that call.
00:05:03 · S1, Donald Muir
[00:05:03] S1: On the topic of customers, I know a lot of founders who work with ARK, and I've never heard anything but really, really good things. And I feel like that's very hard to do as a brand. You know, normally, it's like, it's mixed. The, you know, 10% will be hostile or say they had a bad experience. I've never had a bad experience. So think that says a lot. But also it makes me wonder, was that a hard decision? Like, that can't have been easy to make that call. [00:05:23] S0: I haven't left ARK, at least not in heart or in spirit. Actually, we share office space. Our teams are a 100% in person in San Francisco and New York. We have offices in the Financial District on both coasts. One side is F2. The other side is ARK. I'm on the board. I'm a large shareholder. I see the ARK team every day. So it still feels like one family. Now the teams are focused entirely on their own bookends of the market, but I haven't gone anywhere, and that made the transition easier. Not to mention my cofounder at ARK, who's now CEO. He'd been playing that role for the, you know, call it the last year before I left as I spent more and more of my time building the AI for the lenders. And that was a fairly natural transition. But ARK is always my baby, and 100% of my cash is stored with ARK. I'm still spending tons of time with all the founders, you know, who we bank at ARK and who are, you know, loyal customers of my first company.
00:05:23 · Donald Muir, S1
[00:05:23] S0: I haven't left ARK, at least not in heart or in spirit. Actually, we share office space. Our teams are a 100% in person in San Francisco and New York. We have offices in the Financial District on both coasts. One side is F2. The other side is ARK. I'm on the board. I'm a large shareholder. I see the ARK team every day. So it still feels like one family. Now the teams are focused entirely on their own bookends of the market, but I haven't gone anywhere, and that made the transition easier. Not to mention my cofounder at ARK, who's now CEO. He'd been playing that role for the, you know, call it the last year before I left as I spent more and more of my time building the AI for the lenders. And that was a fairly natural transition. But ARK is always my baby, and 100% of my cash is stored with ARK. I'm still spending tons of time with all the founders, you know, who we bank at ARK and who are, you know, loyal customers of my first company. [00:06:13] S1: So F2 dot AI, saw that you have that domain. What's the story there? I have to imagine that that was not just sitting there on the beam chief ready to buy.
00:06:13 · S1, Donald Muir
[00:06:13] S1: So F2 dot AI, saw that you have that domain. What's the story there? I have to imagine that that was not just sitting there on the beam chief ready to buy. [00:06:21] S0: I went deep in my own pocket for that one. I maxed out all my personal credit cards. I just got my Chase turned back on this week. But I saw the domain. First, the name came to me. Maybe I'll start there. F2. Like I said, I grew up in the bullpen in private equity, where at three in the morning, I would be F2 ing through spreadsheet models, auditing my own work. When your glassy eyed have been staring at a financial model that you've been spending weeks building, you're on V 150. F2 is the keystroke you use to audit calculations in an Excel model. And so in the world of AI, one of the big gaps that we need to bridge, and again, I know this from personal experience, is reliability and trust. LLMs hallucinate. They get stuff wrong. They can't do math. We've developed technology that does not hallucinate, that can do math. We have a category leading Excel agent product. It's core to the platform. F two speaks to that core competency. It's auditability and trust for, you know, the financial buyers, for the private credit and private equity associates who are toiling in the bullpen late at night and long in the weekends. The domain, I scooped up and bought that before incorporating the company. Maxing out all my credit cards to do it.
00:07:31 · S1
[00:07:31] S1: This show is brought to you by Frontlines Media, podcast production studio that helps b two b founders launch, manage, and grow their own podcast. Now if you're a founder, you may be thinking, I don't have time to host a podcast. I've got a company to build. Well, that's exactly what we built our service to do. You show up and host, and we handle literally everything else. To set up a call to discuss launching your own podcast, visit frontlines.io/podcast. Now back to today's episode. And for you, how do you get past that, you know, general perception that AI does hallucinate? So, like, I think every consumer in The US at least is, like, using ChatGPT. We all experience that. How do you sell against that, and how do you make the case that, you know, your AI is not going to hallucinate and, you know, invent shit like the other ones do in our consumer lives?