Episode 109

Roman Schnider from in and on on CFO as a Service

Roman Schnider from in and on on CFO as a Service

What We Discuss With Roman Schnider

Roman Schnider spent 15 years at PwC, where he helped build their blockchain assurance practice back in 2016.

In 2019, he became CFO of the Tezos Foundation, managing one of the largest crypto treasuries in the industry, over $500 million in fiat, BTC, and XTZ.

In 2023, he left to launch inandon, a Swiss firm offering CFO as a Service, accounting, and compliance to web3 companies. And in 2025, he set up Blockfort, a Swiss regulated VASP custodian, built on top of Fireblocks.

Connect with
Roman
Roman Schnider
Partner & Head of Sales @in+on
Connect with

[00:00:00] Roman: CFO as a service is also a bit of a maybe catchy or maybe not, but it's somewhat a brand, right? And Fractional CFO,

[00:00:06] Roman: CFO as a service is actually no more other than finance consulting. 

[00:00:10] Umar: So how would you distinguish between accounting and fractional CFO services? 

[00:00:15] Roman: You have the standardized part, everything around accounting, payments, payroll, VAT, year-end processes.

[00:00:24] Roman: So everything that can be standardized that is regularly ongoing is what we consider accounting. 

[00:00:29] Umar: And CFO as a service? 

[00:00:31] Roman: It's liquidity management, it's product pricing, budget monitoring, and it's preparation for boards. 

[00:00:38] Roman: Sometimes it's pricing of market makers, accounting for loans to market makers, and some have then also the regulatory components.

[00:00:47] Umar: Welcome to The Accountant Quits podcast, where we help accounting and finance professionals learn how to manage a business using crypto.

[00:00:56] Umar: Roman Schneider started his career at PwC Switzerland, where he helped build the firm's blockchain assurance practice in 2016.

[00:01:04] Umar: He then joined the Tezos Foundation as a CFO, managing more than $500 million in crypto treasury.

[00:01:11] Umar: And in 2023, he launched In&On together with Nicolas Hofstetter to offer CFO as a service, accounting, compliance, and payment services.

[00:01:22] Umar: Roman also launched Blockfort, a Swiss-regulated VASP built on Fireblocks covering custody, staking, trading, and wallet services.

[00:01:32] Umar: In this episode with Roman, we explore what CFO as a service means, web3-specific deliverables, lessons from running a foundation treasury, self-custody on Fireblocks versus a regulated custodian like Blockfort, and much more.

[00:01:50] Umar: The Accountant Quits is the official podcast of the Onchain Finance Institute, the leading educational provider for finance teams using digital assets.

[00:02:00] Umar: Their programs, the Crypto Accounting Academy and the Crypto Treasury Management Academy, focus on practical knowledge, including tools required to work with digital assets.

[00:02:11] Umar: Inside their platform, you can also connect with peers working in web3, join focused chat groups, access job opportunities, and attend practical workshop on onchain finance.

[00:02:23] Umar: You can join the community for free by heading to onchainfinanceinstitute.com/community 

[00:02:28] Umar: The link is also in the show notes.

[00:02:31] Umar: Now, enjoy my conversation with Roman.

[00:02:33] Umar: Roman, welcome to the show. It's great to have you on. 

[00:02:37] Roman: Thank you for having me, Umar. It's a pleasure to be here. 

[00:02:39] Umar: So like many of my previous guests on this show, Roman, you started your career at a Big4. In your case, you did 15 years at PwC, mostly in Switzerland as an auditor, and like I said in the intro, you helped build their blockchain and crypto assurance practice back in 2016.

[00:02:59] Umar: In 2019, you then joined the Tezos Foundation as their CFO. 

[00:03:04] Umar: Then in December 2023, you launched In and On together with Nicolas Hofstetter, with whom you worked at PricewaterhouseCoopers.

[00:03:14] Umar: Can you walk us through that journey? And specifically, what do you see from the inside of one of crypto's largest Swiss foundation treasuries that convinced you, "Hey, there's a real gap in the market for a firm like In and On"?

[00:03:27] Roman: I think, you know, when you, when someone tells your CV, you always feel a little bit older than, than you should, should probably feel. And, and one of the funny things is actually, in Switzerland we have something called apprenticeships that you normally start after school. So either you can, can go to kind of higher schools and then go for, for universities, or in my case, you, you do an apprenticeship.

[00:03:50] Roman: And I started with them, so, so 16, 17, and worked as an accountant in a local accounting firm. So actually what I'm doing right now kind of, you know, closes the circle. I don't know whether it's the last step, but it's, um, it's not something I've, I've not done before. And, the journey at PwC, I mean, look, Big4s are a great place to, to get education.

[00:04:11] Roman: I think they have a, a huge resource of, of people and talent and structure. For me, everything that kind of helps you to grow when you're younger. The other thing I loved about Big4s was the ability to travel. Especially, you know, time pre-kids, everything is a bit easier and as an accountant or as a bank clerk in, in the eastern part of Switzerland, you probably wouldn't be able to travel too much, and PwC allowed that.

[00:04:36] Roman: So I was working in St. Gallen initially, then two years during the financial crisis in Germany, and then in Zurich, that's where I did the CPA, and then I spent with my family already three years in Sydney, in Australia, and back here in Zurich again. And that, that was always really, really fascinating and I think with the, experience from abroad, I think especially the one in Germany, I have to say that Germany was an eye-opener at that time because Switzerland is a small market.

[00:05:04] Roman: Sometimes we feel a bit protected. Pressure is not as high as in other countries in the world. And, you come into Germany and you feel like, okay, peer pressure is different. These people actually know how to read contracts even in younger ages, and just being an auditor and tick stuff doesn't work, and so you have to start reading contracts.

[00:05:20] Roman: And at that time it was, you know, CDOs and, mortgage-backed securities and all of that stuff. And it kind of started to, to motivate me to understand more what financial products are out there and why they're built that way and how they work. And that kind of then also fascinated me about DeFi because at the end, you know, a lot of the stuff we saw initially at crypto was at that intersection of regulation and product design.

[00:05:48] Roman: And, being back in Switzerland at that time, after having had the chance to work with companies like Macquarie in Sydney, and obviously on the PwC side, but still, right, and you see these investment banks, you know, you suddenly had the possibility to, to, to see, to see that product and investment banking angle in Switzerland that is traditionally more wealth-based rather than investment banking, right?

[00:06:11] Roman: Sorry, a bit of a long-winded answer, but that kind of brought me into crypto, and that what, what, what made it fascinated to me.

[00:06:18] Roman: And then I'm... I love to build stuff, and I think at the beginning it was really like understanding what are the, the elements that are existing, you know, things like block explorers, can you use the data that you get from, from the first explorers.

[00:06:32] Roman: If you come from PwC world, obviously, are they certified? Can we trust the data? How long do we have to wait until we have finality and stuff like this, and interpret that for a very traditional crowd within PwC. And that, that was really, really exciting. And, and during that time, the Tezos foundation was one of the, the largest clients we had.

[00:06:55] Roman: I was, you know, heading the, the, the onboarding work, then later was on the audit. I signed the audit. And now I could say then I joined them after all because they needed a CFO. The reality was that I tried to make partner in that year, and I didn't make the cut. And, you know, whenever you have these type of moments, then you're like, "Okay, I've been here for 14 years.

[00:07:14] Roman: Maybe I'm not the type for a Partner at PwC, and that's okay." And mentioned that to some of the council members at that time at the foundation, and when I walked back, they were calling me and said, "Yeah, actually, you know, I think we could need someone that helps us to get us a bit more structured."

[00:07:31] Umar: Fantastic. I think Tezos, of course, among the crypto native people, they know what Tezos is. But I think even among people who don't know anything about crypto, Tezos has done a great deal of market awareness. I'm a fan of Manchester United, for example, and I think all Manchester United fans would kind of know what Tezos is now.

[00:07:50] Roman: Yeah. My, my wife was always making the joke that now everybody knows the Tezos logo. The only thing they have to understand is that it's not a toothpaste, but actually a blockchain. But yeah, it was a fantastic time. And I just realized I didn't really answer your question before because you said kind of why, why from that path you then started to think about services that could help people beyond the foundation.

[00:08:11] Roman: I think the beauty at the Tezos foundation was, and for those that are not so familiar with it, I mean, Tezos Foundation in 2017 had a fundraiser, with, obviously the, the, the, the brainchild of, Arthur and Kathleen Breitman, that was incredibly successful. It was very novel. You know, it took staking up for the first time.

[00:08:28] Roman: It was the first staking chain that was there at scale. There were other staking protocols before, but not at that, at that scale. And they raised $231 million. And, uh, one of the reasons, to my understanding is, was that they've been able to enable Bitcoin contributions at that time, which made it successful, besides the fact that the idea was very novel and it was obviously at the right time.

[00:08:52] Roman: So there was always enough money to build a proper finance function. Money was not the thing we had to worry about when building it. But for most of the projects, that's not the reality, right? The reality is that you are, you know, finance is scarce. You have to make sure you run as effectively as possible.

[00:09:09] Roman: And often software is expensive. Service quality, especially in the web3 sector, is not fantastic. And so when Nico and I talked about it, our goal was to say, "Okay, let's build something that is the no-brainer back office solution for those that want to build, that want to build crypto products.

[00:09:29] Roman: For, for them to focus on the product and take the back office pain away." And accounting was one of the first things. You know, I mean, our accounting software costs 50, 70K, annually. It's not a lot, but if you run a business, that's a lot, right? You don't want to have an accounting tool that costs so much.

[00:09:51] Roman: But if you can syndicate it over many, many other clients, that cost becomes 1,000 bucks per client, and that's what we charge for, for our licenses. So deliver the Tezos Foundation experience for smaller clients, I think, by, by scaling it out, by, by allowing a, a modular approach for many. 

[00:10:10] Umar: Perfect.

[00:10:10] Umar: Thanks for sharing, Roman. Now, the title of today's episode is CFO as a Service, and on your website, In and On, you frame it broadly, right? So Fractional CFO, treasury and liquidity management, fundraising, investor relations, everything's delivered as an outsourced partner.

[00:10:29] Umar: Now, the rationale for hiring a fractional CFO is a lot of times because a full-time web3 CFO is expensive, especially at the early stage of that project.

[00:10:40] Umar: So my question is: how do you define the boundary of CFO as a Service at InandOn, and what does a client actually get from you that, you know, they cannot just get from hiring a full-time senior accountant or a bookkeeping firm?

[00:10:54] Roman: You know, CFO as a service is also a bit of a maybe catchy or maybe not, but it's somewhat a brand, right? And Fractional CFO, CFO as a service, is actually no more other than finance consulting. And so the way we often distinguish the two services is you have the standardized part, you know, we'll talk about that later, probably everything around accounting, payments, payroll, VAT, year-end processes.

[00:11:23] Roman: So everything that can be standardized, that can be structured, that is regularly ongoing is what we consider accounting. As soon as it goes beyond that, then it becomes a consulting function, and the CFO title gives you an idea of what are the topics you wanna talk about with that person. You know, it can be that you care about your treasury management, it can be that you talk about actually you wanna understand the profitability of a certain product better.

[00:11:50] Roman: And sometimes those aspects are more like traditional finance aspects, sometimes they're more linked to crypto. But that's how we distinguish it. Is it something that we standardize, where we have the goal to produce it at an as effective as possible rate? We also offer fixed prices. One of the things we saw as well when we started is that many accountants offer hourly rates.

[00:12:10] Roman: They're always very low, and then at the end they realize they have fifteen thousand crypto transactions, and suddenly your accounting costs a lot. Whereas we kind of give you the certainty of a fixed price, you know, that's what you pay. But therefore you also have to help us out in the process of how you deliver, so you wanna standardize it.

[00:12:27] Roman: That's the accounting part.

[00:12:28] Roman: And then CFO part is we have people that have been in those roles, for years, and we, we try to help with all of our knowledge and as specific as possible. That part is obviously less scalable. That, that's us trying to help the, the community 

[00:12:43] Umar: Okay, you mentioned pricing, so, when you put CFO as a service, you know, proposal in front of founders, what are the common objections you hear, and how do you frame that value?

[00:12:56] Umar: And what I mean by that is I think at the beginning of a project's life, I think the CFO as a service is more maybe somehow needed because you're advising them on fundraising, managing their runway, treasury management, and accounting is kind of the function that they're not really thinking at this stage.

[00:13:16] Roman: You know, it's interesting because in some ways I feel like that the CFO is often at the start there, right? That finance accountant. But what he actually does is structuring your business so that the accounting works.

[00:13:28] Roman: So actually, the accounting is the basis for everything. And, and we've seen that especially in the crypto environment, the more you're able to jointly agree on a structure, the more effective.

[00:13:40] Roman: Do you have the numbers at your fingertips? And if you have the numbers available, you can have meaningful discussions without guessing. So many times we see people having spreadsheets that they created themselves. "Hey, that gives me a good idea of what my revenue is." And then you realize it's detached from actual accounting, and so the, the basis is not strong enough, especially when challenge comes in, in a due diligence phase, et cetera, and then you get into difficulties.

[00:14:05] Roman: So, so yeah, I think accounting is a non, not neglectable part of having proper CFO services. And therefore, the first part of CFO services often is to build structures, in our experience, and that can start as simple as wallet inventories. What are actually wallets for a certain company?

[00:14:25] Roman: And what are asset statements, for certain legal entities, legal entities, separate topic, et cetera.

[00:14:30] Roman: And you ask for objections. At the end, consulting is a pretty simple business. You have an hourly rate. In my case, it is significantly lower than the rate I had at PwC. I think that rate in Australia was 1,000 bucks. Here it's, uh, we run on, on Swiss, so it's $360 per hour for me.

[00:14:51] Roman: But that's really for small mandates. So if you have a very small mandate, and the smallest ones we accept is roughly 5K a year, then you'll get a certain amount of hours. You can structure it in the way you want. You know, we can have a monthly call. After the monthly call of an hour, I charge you that hour, and you, you pay that price.

[00:15:09] Roman: When it gets more engaged, what we feel, and, and during the due diligence phase is, it's dedication to this project because accounting is scalable, CFO services are not, and you know, how many mandates can Roman re- really do, and do I get Nico or Roman or whoever I want or Luca and, and would I then get Luca's time or do you give me someone else?

[00:15:31] Roman: And that, that I think is the biggest one, right? Do we get the right expert? Do we get the right availability, the responsiveness? And we also feel that it's interesting during mandates, you know, sometimes when you have a good client relationship, we feel that on the accounting side, we get recommended quite quickly to others.

[00:15:49] Roman: Whereas on the CFO side, people tend not to recommend because you kind of take resources away from you. So I found my person that works for me, so there is no recommendation needed. So, so that's I think the biggest one, you know. Do I get the right person for the right time, and do I have the right responsiveness?

[00:16:06] Roman: And, you know, you, you said it, I mean, we founded it now. We're now in our third year. So, um, and of course, we do everything to, to be as professional as possible and, often it works. Sometimes we have to try harder. 

[00:16:19] Umar: All right. So can we go through some of the deliverables just for our listeners to understand what CFO as a service would entail?

[00:16:27] Umar: Maybe I'm thinking about stuff like, okay, you've got your monthly management accounts, your board pack, your forecast, your treasury pos- position, maybe reporting your KPIs to your investors. What are some of these recurring core tasks that every client would have? 

[00:16:46] Roman: I think recurring is, liquidity management, and this is consulting, right?

[00:16:51] Roman: So it's discussions, uh, based on material produced. It's, uh, liquidity management, it's product pricing, it's, budget monitoring, and yes, I mean, it's, it's preparation for, for boards, that's the management aspect. But sometimes it's boards, sometimes it's investors.

[00:17:09] Roman: And some have then also the regulatory components, so regulatory reportings, et cetera. Yeah. I think these are, these are the most common parts, but, but again, right, it can go m- far beyond that. You know, for in, in a web3 world, sometimes the questions are like, "Look, we're keen to have a trading competition on Binance.

[00:17:29] Roman: Have you done one? Who to contact?" Or, "Hey, we're not reaching Bank A because Bank A, actually, we want a bank account with Bank A. Do you have contacts in Switzerland?" Or, sometimes it's transfer pricing. And that the consulting part is very topic specific and really linked to the actual needs of a business at that point in time.

[00:17:48] Roman: Sometimes it's pricing of market makers, market maker contracts, accounting for loans to market makers. Sometimes it's, et cetera. You, you can feel what I mean. There is a large variety of topics that we, that we don't have to limit. The only thing we obviously say, sometimes we don't have experience in certain matters, and then you flag it.

[00:18:08] Roman: But, that's then the more bespoke part. All 

[00:18:11] Umar: So you did mention some of the web3 specific deliverables there. Let's say there's a web2 CFO listening to us right now, and, you know, he wants to also get into web3. What would you say-- I mean, you did mention a few, but maybe some of the other deliverables that if you have a mandate for a web3 client, you would have to know.

[00:18:34] Umar: You know, things like treasury reporting of your different tokens, understanding onchain transactions, reconciliation, maybe through the accounting software, understanding how to use multisig wallets, how custody works. I mean, we're going to custody a little bit later as well, but what are, you know, some of the main things that they'd have to understand upfront?

[00:18:57] Roman: I, I think the discussion would start what they plan to do in web3. You know, sometimes let's say it's an asset management type or, or a fund management type business. In whatever format they wanna hold a few Bitcoins, then the discussion would be, you know, you mentioned custody, custody, self-custody, and all operational challenges with it.

[00:19:18] Roman: Then you start to talk about block explorer data, whether data tools are needed. You know, sometimes we can start to talk about things like subledgers, et cetera. But really we start with what they want to do. And based on that, you can then start to understand what are frequency of transactions, what are assets held, what venues will be looked at, where would you get the data? What data do you need? And then, yeah, I mean, for the larger ones, how do you prepare it for, for an audit? You know, if you, if you're audited by a PwC to, to stay with that name, you know, and you have crypto assets at Amina, they wanna see a SOC 2 Type 2 or an ISA of your custody.

[00:19:56] Roman: And, and so you have to know what is my selection criterium if I go for a certain custody provider, which is much easier now than a couple of years back, of course, but, um, but still, still a topic to discuss. 

[00:20:09] Umar: Before we continue, we'll take a quick commercial break from our sponsor.

[00:20:14] Umar: Whenever you use cryptocurrencies for your business, the framework for your bookkeeping is a combination of traditional accounting software like QuickBooks, Xero, Oracle NetSuite, and a crypto subledger, which would extract, process, and feed in transactions from wallets, exchanges, and custodians into your accounting software.

[00:20:34] Umar: Kryptos Enterprise is a SOC 2 certified accounting software built for crypto and integrates with 150-plus blockchains, 100-plus exchanges, 50-plus wallets, and over 5,000 DeFi protocols.

[00:20:50] Umar: Using a crypto subledger, you can automatically categorize and track your gas fees, realized gains and losses on trades, and create custom categorization rules.

[00:21:01] Umar: And once categorized, you have the ability to bulk sync your transactions into your accounting software.

[00:21:08] Umar: And with the MCP server, you can now connect your AI assistant like Claude, Cursor, or ChatGPT into your Kryptos account to categorize, reconcile, and sync transactions.

[00:21:20] Umar: If you're still managing crypto on spreadsheets, you're making your life harder than it needs to be.

[00:21:26] Umar: Right now, Kryptos is offering you 10% for the first year of subscription. Check out the link in the description to claim this offer.

[00:21:36] Umar: Now, before we go through what, you guys provided at InandOn, I wanna spend some time on your Tezos foundation years because you joined the company in 2019, so very experienced as a CFO at a web3 company.

[00:21:50] Umar: Now, for the listeners, the market value of Tezos Foundation assets was at 530 million as of December 2025, so it is a very large treasury.

[00:22:01] Umar: Now, Tezos is one of the early ICO projects by back in 2017. It's known as, you know, the pioneer for web3 gaming, NFTs, and you mentioned this earlier, it raised approx 230 million in BTC and ETH.

[00:22:16] Umar: Now, let's say there's a finance professional listening. They've just been handed the treasury function at a, let's say, a foundation or another web3 startup. 

[00:22:25] Umar: So based on what you learned at Tezos, what does a professional web3 foundation finance function looks like across, you know, custody, controls, reporting, and what would you say maybe you got wrong at those early years that maybe today you'd do a bit differently?

[00:22:44] Roman: I mean, I was thinking about what we get wrong, every now and then again. And it's also difficult because obviously the whole sector was kind of built, I would say '17, '16, '17, really to today. So a lot of the stuff you did at that time was also driven by the availability of tools you had.

[00:23:05] Roman: And so, you know, in hindsight, you're also smart. So it's a bit hard, but I try to answer it. I, I think again, right, even with large endeavors, at the end, the first step is to understand what is your first-- what, what is your balance? What is your data availability?

[00:23:19] Roman: In the sense like what are your assets? What are the places where you store them? Are you sure that that's the asset reporting, that the asset reportings you get are the right ones, you get the right values? Then you start to think about asset statements, liquidity statements, you know, where you have the assets, in different in nature.

[00:23:36] Roman: You probably sort them for liquidity reasons. You start to think about where you store them. When you start to think about storage, you start to think about counterparty risk, you know, different legal jurisdictions in the case of crypto. You know, sometimes in Switzerland, we are lucky, you know, we have a crypto, law or, a DLT law that, that, that clearly defines that segregated assets per client are outside, bankruptcy proceedings, so their credit risk remote.

[00:24:03] Roman: In other situations that may not be the case. So, so this is what you kind of start to, to think about.

[00:24:07] Roman: And that's not our approach. You know, people would do it differently. I mean, I'd like to start simple. You start with policies that make sense. We had a council. So the council is the board of a foundation in Switzerland.

[00:24:19] Roman: The council existed first, and, the first employee was the President of the council, so in like a dual operational role, and I was like the, the CFO COO, which, you know, when you have not many people doing jobs, C titles are very easy, right? It's like, what's-- there's just one, two people. And, and you, you try to understand what the role is of, of the vehicle and, and the finance part is one part.

[00:24:44] Roman: You know, you need to have your accounting tools. You need to have an understanding, again, depending on size, whether you need certain accounting standards. In Switzerland, normally you have Swiss code, but if you are above a certain level, you need, a recognized accounting standard. In our situation, that was then IFRS.

[00:24:59] Roman: We chose IFRS for SME as there is an IFRS standard for small and mid-cap. And the main reason was that there you, you could avoid the IFRS 9, or the IAS and IFRS 9 disclosures that were literally undoable at that point in time for crypto assets. When you wanted to be audited by PwC, which in our case was a more or less mandatory thing, the Big4 audit.

[00:25:20] Roman: So, building that structure from what are limits, you know, who approves what, what is management approval, what is board approval, how do you make sure that you have segregated duties so that not someone can access stuff alone. It's, it's very trivial stuff at the beginning, but that's where the crypto, crypto suit...

[00:25:41] Roman: And then-- And that's the, that's what hindsight makes it difficult. At the beginning, a lot of stuff was done for adoption. So for example, if the Tezos Foundation onboarded significantly more service provider it needed, and the reason was that every onboarding was an adoption, right? If you onboard with us, you also have to offer Tez, for example.

[00:26:00] Roman: If you onboard with us, you have to offer our tokens or our token standards. So i-i-it went beyond just the business aspect, and it had this adoption element of we try to win as a protocol. And we try to have that large community to win that community race. And I think that's also where, in hindsight, different strategies would have been possible.

[00:26:24] Roman: So, so I think, you know, whether at, at some points, people came to us and said, "Hey, you really do a great thing having so many grantees because you build a community and everybody can participate." The reality was it obviously... It, it, it came with an inherent chaos and an extremely heavy coordination attempt.

[00:26:42] Roman: Others have tried one, two things and those... And, and, and, and brought them to, to financial success. So a- again, right? The foundation has a very specific goal of what it builds, and, and we m- we optimized for it. We certainly did things wrong. I mean, at that time, I was really proud that we never lost assets.

[00:27:03] Roman: You know, initially there was a lot of self-custody situations, bakers or validators very early on. We started early on with Cryptio at that time. Antoine didn't really have a lot of funding rounds when he started, and I think his LIFO, FIFO stuff wasn't, uh, finalized entirely. I think some of our employees helped there.

[00:27:23] Roman: And so yeah, it, it was really that building that basic infrastructure and at the same time trying to convince others that your protocol actually has an impact. 

[00:27:32] Umar: , I'd like to move on to a follow-up question on the Swiss foundation specifics. We've covered Cayman and Panama foundations extens- extensively on this podcast, but from the inside, what's distinct about running treasury and governance under Swiss foundation, their supervisory regime, the audit expectation there that maybe founders choosing Switzerland should understand?

[00:27:59] Roman: I mean, you know I'm an accountant, I'm not a lawyer, and so I, I would probably... And, and especially, right, for example, Panama foundations I don't really know a lot about. So I don't-- I, I can't directly compare, but I mean what I always found fascinating is the reason why we ended up with foundations was a regulatory quirk, right?

[00:28:14] Roman: People tried to understand how they can run ICOs, fundraisers without being considered a bank or requiring regulation. And, you know, if you accept donations, then that's okay. If you can have a purpose for them to be used for, but if then, if you have a setup that looks like a deposit, then suddenly you become a bank, right?

[00:28:36] Roman: So I think that's initially, and I believe it was around Ethereum, that was why Zug created the foundation model that was very, very fashionable for quite some time. I think some of the law firms have also leveraged that legal contract quite a bit in order to make, to make their benefits. And I think that's okay, right?

[00:28:53] Roman: That, that was a model that was used. And regulation depends on size. So smaller foundations, I believe, and I'm guessing a bit, are regulated by-- overseen by the cantons, so the, the Swiss states, if you want so. And then the larger foundations, like the foundation from the, from Tezos, were supervised by the, eESA, which is the, the Swiss foundation supervisory authority.

[00:29:17] Roman: Yeah. You know, sometimes for those that are in banking, this is not a FINMA type regulation. FINMA is a Swiss banking regulator, so it's not and, and I don't mean that in any disrespect, but it's not like the same active regulation or regulatory communication you'd have in a foundation.

[00:29:33] Roman: It is much more that you do report annually on your activities, and then the authority obviously dis-discusses your report, may come back with question, asks for, explanations or certain deliverables. So that's-- you, you would feel an annual reporting. The lighter version of it is an association. So Switzerland is full of associations, and it's a model that has then taken off for some time when people felt like the foundation setup is too cumbersome.

[00:30:05] Roman: And that, that works relatively similar, but the supervision model is, is, is not, not there. So that's really like a, a lighter form of a foundation. I think for us, the biggest myth buster was, you know, when you look at the Tezos foundation structure, you have Tezos Foundation as a foundation, and underneath it there is Tezos AG, so a Swiss company, controlled by shares.

[00:30:25] Roman: And initially, my understanding was it was created because there was uncertainty whether a foundation can actually do operational business within the foundation. And that's, you know, people found out that this is easily possible, so most foundations run the business in the foundation. And so, that AG subsidiary that some had was not really necessary.

[00:30:47] Umar: All right. So moving on to InandOn, the company that you've launched at the end of 2023, like I mentioned earlier. So going from being employed as a CFO with a defined mandate to now building a firm from scratch, it's a completely different muscle. You now own prospecting, hiring, services, everything at once.

[00:31:08] Umar: So my question is twofold: How did you and Nicolas, your Co-founder, maybe divide responsibilities, and how did you land your earliest clients? 

[00:31:18] Roman: Yeah, I think the advantage-- So I mean, for me, I always had this condition, which is I don't wanna create a company myself. Because I also feel like having a team for me is really healthy.

[00:31:31] Roman: Having someone that, that challenges me when I have two crazy ideas or two boring ideas. So to have Nico, who works very differently and works for, for different people, um, and much better than I do, there was somewhat a natural, allocation of tasks. Nico is much better with everything around hiring people, people skills.

[00:31:51] Roman: I think he listens much ca- more careful and understands where, w- how people would fit together. Whereas, I do like building operations and businesses themselves, so also the BD part. And, we knew that we will have two mandates to start with. That gave us the certainty to sit together in a room.

[00:32:12] Roman: I remember I started first. Nico was still finishing his job at that point in time. And so I was sitting in actually, one of the law firms here in Switzerland called Lexer. They're also doing a lot in the, crypto space, and they sub-rented an office or a, a room for us. And I think one of the first clients was also from them, where they said, "Hey, we have a client that needs accounting.

[00:32:33] Roman: Do you guys wanna look at it?" And then as it always works, "Hey, I heard you guys are building an accounting function for web3. I do have a web3 client that needs help." And, and, and that's how you start to, to, to, you know, write your first engagement letter, figure out that you need terms and conditions, and, uh yeah, you create some things yourself, copy other things, and then, you know, try to make it a business, I guess.

[00:32:55] Roman: And, and then over time you realize, okay, now, you know, some s- some stuff has to be more professional and, and, you build it. So I think having had the chance to be part of the build-out of the Tezos Foundation gave us this safe environment to build, to learn how building a company could look like, without having the economic pressure that many of us have.

[00:33:18] Roman: Now we have the economic pressure and, it's still interesting. It is still very, very different to, to building a foundation that is not your, your, um, you know, your two clients, and you worry in the evening because you may lose one and then you have only one left at the beginning.

[00:33:35] Umar: I speak to a few professionals in the industry. You know, they worked for four, five years at a web3 company as a CFO, Head of Finance, and they're actually also now taking the leap to launch their own fractional CFO firm. It's good news for the industry if we can have more people to come and service this industry.

[00:33:54] Umar: What have been maybe some of the hardest lessons in building and training the team around operating, you know, crypto? You have a lot of experience in crypto but then hiring and training people is quite something else. 

[00:34:08] Roman: Yeah, and you tr- obviously-- I mean, one thing we see is that often... So there is one man, one woman, one man services, that is hard, right?

[00:34:18] Roman: Because the work has different complexities, so you would wanna structure them over a larger team that has that ability to take on these different levels. But also you need, a replacement, otherwise you burn through your energy because otherwi- it's always just you, that's hard. And so in some way, we also offer, you know, that, that ability for people to join us.

[00:34:42] Roman: So we've just now had a CEO/ CFO joining us very recently that, uh, was with the bank before, and he intends to, to, to become more of a fractional CFO. But in-- and the reason why they normally join us is that they say like, "Look, we don't wanna go through the hustle of creating everything ourselves."

[00:35:00] Roman: Then the team aspect, right? How can we make sure we can also offload and have someone that challenges us so that we're not alone, but still in a relatively small, small environment. I think You know, I, educating, I, I like to educate, but I'm also aware that I'm not always the most structured guy in educating.

[00:35:19] Roman: Again, right, that's something that Nico is much better at, for example. We've worked-- we started with a team initially that we knew well, that also knew crypto well, so we didn't have to start with education at first. Then we, we very soon we got a lot of pressure on the cost side for the basic accounting matters like accounts receivables, payables, credit cards, and we, we had to think about offshore locations, just because of the cost level here in Switzerland being far too high.

[00:35:48] Roman: And we thought about, you know, the, the obvious candidates in Europe. Our goal was that it had to stay close. We had to be in, in good contact, and we wanted us to be a team, even if we would have a, a near-shore, with us. So we ended up in Lisbon as well, mainly because we tried vendors to help us.

[00:36:05] Roman: We realized if we wanna get the team that we like to work with, we have to find the team we like to work with. And then Nico lived a year in Lisbon and, and s- and started hiring from scratch, and we built the same office we have here in Zurich, also in Lisbon, so that it's like the same, the same for both teams, and we have regular...

[00:36:21] Roman: So that-- I mean, you're right, that is certainly the hardest part, finding team and then also moving with the team as a second thing. Like, consulting is easy in some way, right? You, you get a call, you do a couple of hours, you get charged, and the delivery risk is the biggest one, I guess. But it's easy in the sense that there is no process or optimization needed.

[00:36:43] Roman: At the maximum, you have to write the report, or otherwise it's literally just hours, hours reporting. Whereas accounting is very different, right? It's a very different beast. In order for it to be a little bit successful, you know, it's a thin margin, margin business anyway. In order to be a little bit successful, you really have to understand how you, how you work through those processes, and I would lie if I would say we're 100% there, everything is fine.

[00:37:06] Roman: So, that still takes a lot of our time. You know, how do we, how do we structure it? Especially in crypto as well, especially because we wanna be fin-fintech tech, so not just crypto. And so just going for QuickBooks is not a solution for us, but maybe having Abacus in our case has not the same dynamics that we would have see with other tools.

[00:37:26] Roman: It does offer many other advantages, but, but not that part. And so, you know, working on that process part and managing the teams, and that's a, a big thing that goes together because the team is also frustrated because they're good, good people, right? If, if you don't move fast, fast enough, they'll try to find solutions themselves.

[00:37:45] Roman: And that's great, right? Because they, they wanna build the company with you. But at the same time, it's not great because everybody builds its own process, and you have to make sure that everything looks the same, because otherwise it doesn't become that sausage, sausage factory you need. I think it's basic stuff of building a scalable business, but for me, that's really hard.

[00:38:02] Umar: And could you provide the listeners with an overview of the different service line that you're delivering? I think everyone understood that you provide CFO as a service, also accounting, but then I noted you also offer compliance services and payment services, which we'll also discuss in one of the next question.

[00:38:20] Roman: Yeah. As everything, you know, things grew, I mean, you, you remember the beginning, our thesis was to build the back office that, you know, focus on your growth. You know, we want you to focus on your business. We take away the pain, the back office pain. And a lot of back office pains comes with integration, with access, you know, being it banks, custody providers, et cetera.

[00:38:43] Roman: And so you have the accounting-- on the finance side, you have the accounting part, which is the, the basic, you know, that's accounting, payroll, taxes, it's also payments. And then you have the consulting part on top of it, which we discussed as CFO services. And then you have the compliance part that is often very closely related for many, right?

[00:39:02] Roman: It's the, the, the money laundering aspect. For example, if you have some form of, an AMLA license, so everything around money laundering, if your license is in a lower type category, which comes with certain compliance tasks that feel a bit similar to on the accounting side. It's just on the legal side.

[00:39:19] Roman: You have the management of this, of the accounting aspects of a business and growing a business, and then you have the consulting of the, the regulatory part as well. You know, where would you wanna have a license? How do you respond to a regulator? How do you structure a certain product, et cetera?

[00:39:35] Roman: That part comes close. And then the payments part, and that's then why we kind of created Blockfort. Again, I mean, the payments part is when you do CFO services, one of the things we've seen is that crypto people tend to be pretty trustworthy, or maybe they just thought that these guys sit in Switzerland and are okay, and then you kind of get the wallets and like, you know, now you can manage my assets.

[00:39:56] Roman: And that's, you know, we appreciate the trust, but that's not easy, right? You can't do that officially. You, you kind of can do it in a management function, but it's a very different beast, right? You could lose suddenly, you know, you pay a certain amount of money and you-- that could... especially in crypto, you know, these things can be loose.

[00:40:12] Roman: It needs different processes, different T&Cs, different agreements. And in Switzerland, actually for us, it needed a, an SRO license. So, um, so we are a-- we have an entity that's a VQF licensed member that is, a financial intermediary, so we can make payments, custody, escrow setups on off-ramps and trades.

[00:40:35] Roman: So we're luckily-- we have a little bit of a custody, that, helps us, I think, to be agile for our clients. You know, if you think about that suite of back office services, sometimes you need cash. Maybe for you it's harder to do all the efforts of getting one, two, three bank accounts. If you can leverage that setup, if you can leverage the Fireblocks solution, if you can leverage the Elliptic part, if you can leverage the onboarding tools.

[00:40:59] Roman: It's the same story as with accounting, right? You have one block of software. If multiple people use it, it's a benefit for the cost. 

[00:41:07] Umar: Before we continue, let's take a quick commercial break from our sponsor. If you're serious about running your business using crypto, you need tools built for crypto.

[00:41:17] Umar: One of my favorite apps, and yes, one that I use multiple times a month, is Request Finance.

[00:41:23] Umar: This app has been a game changer in how I invoice my clients and receive stablecoins and crypto directly in my wallet.

[00:41:31] Umar: But hey, there's so much more to it.

[00:41:34] Umar: Request Finance is the all-in-one platform for crypto operations, accounts payable, accounts receivable, expenses, and even accounting.

[00:41:44] Umar: Plus, they've got a killer multi-payee off-ramp to over 190 countries, meaning you can pay someone with crypto, and they get fiat in their bank account.

[00:41:55] Umar: Request Finance connects to over 20 blockchain networks, including Ethereum, Solana, Base, Arbitrum, Near, and more.

[00:42:04] Umar: Here are some features that I love.

[00:42:06] Umar: Batch payments using your preferred wallet of choice. A single dashboard to track all invoices with real-time updates. All invoices contain the transaction hash for easy audits.

[00:42:20] Umar: But hey, don't take my word for it. Try it out for yourself.

[00:42:23] Umar: Head over to requestfinance/partners/theaccountantquits and get two months for free with all the pro account features.

[00:42:33] Umar: Join 3,000 finance leaders today using Request Finance and make crypto operations simpler, compliant, and less stressful.

[00:42:46] Umar: We'll go through that specific solution, Blockfort, in a bit. But I, I had a follow-up question. In terms of, like, compliance services, what would be those specific services? Would it be, like, maybe obtaining a specific license and you, you assisting them with that, for example? 

[00:43:02] Roman: I mean, and obviously there is an overlap to legal firms.

[00:43:04] Roman: I mean, it can be everything from, you know, someone changing something in the corporate register, to changing certain regulations and policies. Often it has to do with money laundering activities, so anti-money laundering activities, you know, in the sense of checking, checking payments, checking, onboarding clients and, and, and helping with the documentation there.

[00:43:28] Roman: Report to regulators, reporting to regulators, responses to regulators. And, and then, you know, that, that kind of goes hand-in-hand with then licensing, et cetera. And again, you know, we have a very, very kind of like thin or, or small vertical in that web3 and tech space, and I think we're really good there.

[00:43:48] Roman: And so our decision was to be broader-- a broader service partner for that sector to be really a, you know, a service from one, from one trusted party. But as a consequence, we, you know, we would not accept clients that are outside that sector because we've-- we don't see our value add there. 

[00:44:06] Umar: Yeah. now in January of 2025, roughly a year later after launching InandOn, you set up Blockfort.

[00:44:14] Umar: Blockfort is a separate entity. It's a Swiss-regulated crypto custodian. Providing custody, staking, and trading, and it's built on top of Fireblocks. And you're also currently the CEO of Blockfort. So when you provide accounting and those fiduciary services, there are independence and licensing reason you generally can also take custody of your clients, right, under the same roof.

[00:44:39] Umar: Can you walk us through exactly why InandOn couldn't offer what Blockfort offers and maybe what those licensing, licensing constraints were, and how do you think about the wall between the two businesses? 

[00:44:52] Roman: Yeah, I mean, you know, you talked about the segregation between Nico and myself. I mean, at the moment, Nico is the CEO of InandOn, and I'm the CEO of Blockfort.

[00:45:01] Roman: So I run Blockfort as a business predominantly, and Nico runs InandOn. So, Blockfort is a virtual asset service provider under the Swiss regulatory framework, so it's a VASP, and it's a Swiss custodian. As you said, you know, it can, offer storing, exchange, payment services for non-securities.

[00:45:24] Roman: It's name-- It's not a bank. It's, it's, it's a crypto custodian that has access. But what it actually does, and then we're getting back to what we discussed on the finance side, it, it purchases expensive software and it makes them available to many. And for that, it gives a price advantage to those that want to offer it.

[00:45:45] Roman: What I mean is Fireblocks costs you, I'm making that up now, let's say 40, 50K a year. Elliptic, Elliptic costs you 10K a year or 8K a year. Sumsub costs you seven, 8K a year, et cetera. So you , you probably want to have a setup that is SOC 2 Type 2 controlled. So for us, the starting point really was in after year one, look, we need to get cheaper in accounting.

[00:46:11] Roman: That's when we start to think about the, the offshoring or the nearshore in solution in Lisbon, and we want to continue to be enabled to make-- to help our our clients also on the, on the crypto side, which is really like the, the main businesses are operational payments, you know, that's the regular payments to c-contractors.

[00:46:30] Roman: And maybe you can talk about that a bit more because it gives you an idea of how the segregation works, custody and escrow mandates. So for example, if you have OTC deals between two parties and you want to have someone that secures the fund so that they get only exchanged when both, both sides are, are met.

[00:46:45] Roman: But if you look at payments, for example, the, the creation of a payment file is a traditional finance function, right? So accounting would probably offer a mailbox or a, a certain logic of how to collate data. There will be a process of how that data will be aggregated in a certain cadence, and then payment files will be going through an approval process, and that-- but that's all InandOn, right?

[00:47:09] Roman: That's accounting work.

[00:47:10] Roman: But at one point in time, someone actually has to pay those invoices, and, and that, is something if you do it with, with crypto, that is something then that Blockfort would do. So Blockfort would make the payments, to those parties, obviously under the framework of a VASP in Switzerland, which means, you know, onboarding of clients, means travel rule, etc.

[00:47:34] Roman: And, and again, right, we-- if you would ask us whether we wanna go out there and become the next huge custody service, that is actually not what we wanna do at all. But we wanna be that service provider for crypto companies that really gives you an end-to-end service if needed and makes you very agile and fast, and that's what our clients love.

[00:47:55] Roman: We, we do see that many of our clients on the accounting side love to be part of Blockfort as well because it gives you the-- it gives them the speed. And it also reduces the pain of onboarding because if data sharing, sharing is allowed, if they're on our accounting side, we do know a lot of the data already, so the onboarding is easier, and you don't have to start from scratch, and, and that, that's also seen as a benefit.

[00:48:16] Umar: So to clarify for the listeners, Blockfort is required because as an accounting firm InandOn, so you cannot make payroll, payment services, as I understand it from InandOn, and it has to be done through Blockfort. Is my understanding right or, uh, are there 

[00:48:34] Roman: s- It's always, you know-- There, there is always gray zones, right?

[00:48:36] Roman: I think you could make a point that if you have a management function with a client, so if I'm acting as a CFO for client A, then I'm allowed to manage the facilities of that client A, and that may include their crypto wallets and their crypto exchange, etc. But I mean, the-- for us, we felt like if we wanna offer it as a, as a product because in Switzerland it's clearly defined that if you touch the payments processes in crypto, you are under the financial intermediaries regulation, which means that you need an SRO license.

[00:49:13] Roman: And for us, then you get to the question, you know, do you want to do it right or not? And if you want to do it right, then I think creating something like Blockfort isn't just having the license to make the payments. It also forces you to think about your custody setup, to, to buy the right infrastructure.

[00:49:31] Roman: Yes, we're using Fireblocks, for example. And you know, you have to think about how to do your key ceremonies in a way that, in our case, PwC can audit, so you get a ISAE3402. And it forces you to have the right compliance tool in place, etc. So it gives you a lot of... And, sorry, I forgot that, you can also define the contracts much more precise for the business.

[00:49:54] Roman: Now you're in a payments business, you're in a custody business. The risks are different. The terms and conditions are different. And you wouldn't really have that on the InandOn side, right? That's accounting. So, so it allowed us to segregate the different businesses, keep them with enough proximity to not lose the value why we created it initially, and we get the reduced risk from having had the chance to build a service that is distinct, a framework that is distinct to that service, if that, if that makes sense.

[00:50:26] Roman: So again, right, it's the end-to-end, end-to-end service for our clients, but at the same time, you know, we don't want to... You know, it's, it's crypto assets. There is risk linked to it. We want to make sure we do it properly and the client understands the risk. 

[00:50:40] Umar: So a lot of our listeners are already using Fireblocks.

[00:50:44] Umar: What I'd like them to understand is why should they use Blockfort? Why can't they just use Fireblocks? And do they have to be based in Switzerland to use Blockfort? And what, what's the specific value proposition of Blockfort? 

[00:50:59] Roman: And I obviously don't know all of your listeners as good as you do, Umar, but what we've seen is most often clients held whatever wallets for the self-custody part, and then they had exchanges or custody relations where they held assets.

[00:51:14] Roman: Custody already for the larger ones. The, the small to medium size often had their Metamask wallets, maybe Safe wallets, maybe they had Ledgers. But, but very often they had also Metamask setups. And it was also quite normal that when we onboard the clients, there was the stories of lost funds. Like, oh, you know, if, yeah, that happened, there was a hack, or we lost the access to that wallet, or, oh, yeah, we have a Safe, but there is a key no longer in our control.

[00:51:38] Roman: And it's not something that I... And I'm not pointing finger to these, these, these clients bec- the reality is that they wanna focus on their product, and they wanna grow it, and that takes a lot of their mind space. And create a key in a meaningful, secure setup is a, a mystery in itself.

[00:51:55] Roman: Like doing a key ceremony in a way that is auditable, having a run book, having, having your hardware set up in a way that, that makes sure that you're not actually... you're absolutely sure that no one was able to, etc. So to answer your question, I think if you have a wallet that you just use for retail payments, you know, use the wallet for retail payments, there's no need for Blockfort.

[00:52:18] Roman: If you actually go for a Fireblocks instance, the s- the things start to change. You go for a cost. And yes, of course, you can run your key ceremony with a simple plugin of Fireblocks, and you can have a storage as well. But, I mean, unless you want to spend a lot of time in, in cybersecurity and, and IT setups, it's not your core business, you know?

[00:52:39] Roman: Why should you think about updating your devices regularly, having fresh devices the way you do the key ceremonies, having regular controls around your processes? Again, your business is to sell your products A, B, and C. And for us, it's the intersection of how do I get as much value as possible for as little money as possible, then you get to the managed part.

[00:53:01] Roman: And, and I mean, you know, I wanna come back to the CFO discussion because I find that also interesting. Very often you hear like, "Hey, I wanna get a Fractional CFO because a CFO in crypto costs 300K." Making up that number, right? I-- That's a very high number. But actually the number is higher because if you wanna have a finance setup that gives you a sustainable setup, you probably talk about, depending on which-- in Switzerland, you talk about, you know, three-quarter of a million to a million because you need three to four people, you need policies, you need infrastructure, you need tools, you need replacement logics.

[00:53:35] Roman: You need to spend the time to hire them. You need to spend the time to educate them. And so for us, really where we see the benefit is in the CFO part, as soon as you are below $1 million a year for your finance team, that's when we start to be economically more valuable than, than for the other setup.

[00:53:52] Roman: The same applies on the Blockfort side, you know, comparing Fireblocks. I... It was a bit of a long-winded answer. I hope, um, it was clear to the listeners what I meant. 

[00:54:02] Umar: Got it. So the onboarding experience would be very similar to Fireblocks.

[00:54:07] Umar: I mean, it's built on top of Fireblocks, but you're taking away all those-- the time that they need to set it up, 

[00:54:15] Roman: The onboarding would be different because- 

[00:54:16] Umar: Oh, 

[00:54:16] Roman: the onboarding- You don't, you don't onboard with a tech tool, you onboard with a service provider. So really what you do is a KYB if you're a business.

[00:54:23] Roman: So I know your-- know your bu- know your business, so you, you have to disclose ownership, legal structure, et cetera, and, you know, source of wealth and source of funds have to be assessed for the businesses. So that it's similar to, for example, onboarding with Kraken, for example. It's a service provider you don't have to care about the Fireblocks part because, um, there is an online, front end that you can use to see, manage your balances.

[00:54:50] Roman: So the, the benefit really comes from having access to a Fireblocks without being obliged to make sure that your Fireblocks is safe. And I mean, nowadays we see it. I mean, at the moment we have nearly weekly discussions about hacks because of AI, and, a lot of these hacks are not very complex hacks, but they get more and more difficult to avoid.

[00:55:13] Roman: I mean, you know, we have situations where people, you know, break into, clouds, and, uh, situations where, where, where you, where you talk about bounties on the, on the protocol sides much more often than before. So, so that security aspect has become much more expensive and yeah, our option allows us, A, to share, to share the cost amongst the group, and B, it gives you that service with a proximity to an accountant and a compliance person, so you have that backup service.

[00:55:43] Roman: It feels a bit like your back, your back office without being your back office. 

[00:55:47] Umar: Perfect. That's clear, Roman. And I'll, of course, also include the link of Blockfort in the show notes for the listeners. Now, I'm looking at the time right now, so I wanna be respectful of your time. We are nearly hitting the hour mark, so probably there's a last topic which I wanna go through, which is on using AI.

[00:56:07] Umar: I'm trying to make it a point to touch on AI now in every episode. So you've been an auditor, a foundation CFO, and now you're running both an outsource finance firm and a custodial solution. Now, when it comes to AI, the kind of recurring tension I hear from a lot of accounting firms is they don't want to put client figures into an LLM as their confidentiality won't allow it.

[00:56:33] Umar: But that caution has an opportunity cost. The firm won't touch... I mean, the firm that doesn't touch AI kind of falls behind the one that figures out how to use it safely, right? So how should a firm like yours InandOn navigate using AI for clients? Are you actually already using it? 

[00:56:52] Roman: Yeah. I mean, it is a question you don't really know where to start with.

[00:56:56] Roman: I mean, the, the speed of AI is tremendous, and it's fascinating. We feel it in different areas. I mean, for example, I mentioned before that we have a front end for Blockfort. I'm pretty sure that we would have been spending, you know, three or four or five times the cost of creating that, that solution, a couple of months back.

[00:57:15] Roman: And, and so that these things are in-- Or we started to look at reconciliation. So they... Yes, you get into data protection discussions. What is interesting from an accounting services provider is that your clients', data sensitivities may not be the same. So sometimes clients tell you, "You know, we don't care if people know this and this information."

[00:57:35] Roman: Then you obviously have to, have a discussion why, why you care in certain circumstances. But the way we started doing it, and again, Umar, I think this is a topic where Nico probably better than I am, um, I think I'm beginning to gray hair for that. But I remember maybe nine, 10 months ago, we had a longer session of how could we use AI for, for InandOn, and the main reason for the discussion was we wanted to have some form of AI strategy.

[00:58:02] Roman: In my mind, it's always nice to have some form of logic of how you roll it out. Realizing that-- And we realized that that strategy is very hard to grasp. So rather than a strategy, we, we've been lucky that we have one colleague that is really good in AI. He's, uh, he's one of our IT, IT members, um, and he took on the role of a, an AI coach.

[00:58:25] Roman: So he created a first environment and created carefully. Now, right, he created the, the, the combination of... he, he thinks about the models and model combinations that are necessary in order to go for certain tasks. So in, in other words, we kind of tell him what we could need from a business perspective, and he will build it for us.

[00:58:44] Roman: And we have a platform that is available to us. We can log in, and use certain, um, bespoke applications. And one of the focus areas first was public data. We spent a lot of time, you know, we use TRES, we use Cryptio, we use Breezing, all great tools, but one of the things we always had to think about is data consistency.

[00:59:04] Roman: You have to do those checks. With AI, they're much, much simpler. So those reconciliation tools that give you the gaps, that actually tell you, "Hey, you know, I have a gap here, but actually I think if you would add that token, the gap would go. Do you want me to add the token and do the rec again?" "Yes, okay, I want you to."

[00:59:18] Roman: Okay, and then it learns it. And so all of these aspects actually, that are based on public data work, public data work that, that we could solve very successfully, in, in each- in initial stages. But, but yeah, I mean, it is a blessing and a curse, right? I had a discussion today, with the team because we actually thought about how to give access to everyone to some, I think in that case, Claude, Claude models and The problem is not the cost per se, you know, but if you give access to, to, to an AI, what is the mandate you give to employees?

[00:59:53] Roman: You know, I mean, if I'm employed and I get access, you know, now I can start to build my own software to do my job. But do I want a employee to build their own software? And, and if, if they do it, then how do I make sure that someone else has to, y- you know, that it kind of is linked again, you know, coming back to the processes in, in the, in the highly automated process, that is actually not good, right?

[01:00:17] Roman: Even though it's a good solution for that person, it creates individual solutions that will ha- that will avoid you your automation, benefits. Whereas on the consulting side, obviously it's, it's, it's very helpful. So you can hear we are still, um, seeking the perfect solution. I think we're blessed with good talent.

[01:00:34] Roman: Um, but yeah, I don't think we can say we found the solution. We try to grow with it and it helps in many areas. Yeah. 

[01:00:42] Umar: Yeah, I understand why it's problematic. Probably really using it to set up workflows, processes, but having defined areas and yeah, you have control on how every team member is actually using it.

[01:00:55] Umar: Perfect, Roman. There's a last question which, uh, I want to go through before we wrap up is actually on looking ahead. So there's an en- enormous amount happening right now on the infrastructure layer that a lot of your clients are touching, you know, stablecoins moving more and more into payment rails, tokenization of real-world assets, hot topic for the year, and onchain vaults as well.

[01:01:20] Umar: So across all of that, where are you maybe seeing the biggest uptick in client demand right now? Which of these themes are you personally most excited to build over the next couple of years? 

[01:01:34] Roman: I mean, I think, you know, we all know that crypto at the moment is in a more bearish environment and, we see growth, um, that is still coming from a, the stablecoin sector with, I, I guess a lot of it driven by clearer regulation.

[01:01:47] Roman: In Europe, it was MiCA, for example, with the EMIs, for example. That helps, right? It helps service providers, and it, um, is, is, is one of the, the real use cases. We do see others, I, you know, I think all of these, these times of compression, you know, these, these... when you have breathing organisms, you know, sometimes you're bullish and the things is hot, and you do and produce a lot of things, but you have no time to digest. When it contracts, it's normally the time when you think about your structure and your organization.

[01:02:16] Roman: So we see a lot of efforts going in there, and those are also very interesting for us. You know, people creating new licenses, but also people ensuring that they have their, their management, their, their numbers, their monitoring, their dashboards under control. We see a lot of that. And then, yeah, I mean, especially in crypto, I mean, we also hear a lot of the, the uncertainty, I would say, for the lack of a better word.

[01:02:39] Roman: You know, I mean, I guess the Tezos Foundation has invested millions and millions of dollar in tech. I don't know whether that tech could be created by AIs in a few minutes in the near future. So, so that, that whole change impacts, and I think many people try to get their head around how they actually make business given these new circumstances 

[01:02:59] Umar: Perfect, Roman.

[01:03:00] Umar: Thanks for sharing. There's a last question which I usually like to ask to my guests before they leave is, do you have a favorite quote or a maxim that you live by? Something maybe that you repeatedly repeat to yourself maybe? 

[01:03:12] Roman: I, traditional ways, way of phrasing it, phrasing it would be step by step. What we often said funnily in crypto is block by block.

[01:03:20] Roman: So it's like, yeah, I mean, you know, I, I, I'm not blessed with, incredible intelligence. I think what I can achieve is with others and with, with constant work, and, and, and that's how we really... And that's not a maxim, but obviously we, we wanna we're blessed to be able to work with people that we really like and I think that's also a fantastic environment and a value we get every day.

[01:03:43] Umar: Perfect, Roman. If people want to reach out to you, Are you active on socials maybe, or what's the best way to reach out to you? Of course, I'll share, I'll be sharing all the website, of InandOn as well. 

[01:03:55] Roman: Yeah I think, you know, during some times at, uh, as a president at the Tezos Foundation, some of the social- socials were too much, and so I reduced it a bit.

[01:04:05] Roman: So you won't find me on Reddit. Sorry, guys. But yeah, I'm on LinkedIn. and obviously you see the contact, emails on the In&On website, or on, on, on Blockfault. yeah, Please, please reach out. You know, we're, we're here in Zurich, and we're often in London or in Lisbon, of course, right?

[01:04:20] Roman: So, um, yeah. I'm, I'm always open to get beers on a beach bar in Lisbon. 

[01:04:27] Umar: And for people listening, the website is inandon.ch if you wanna check it out. Fantastic, Roman. Thanks a lot for joining us today. It was, yeah, it was a pleasure to have you on to discuss CFO as a Service in more detail, and, let's keep in touch.

[01:04:45] Roman: Perfect. Yeah, no, thank you very much for having me, Umar, and I think it's a really great effort that you, you know, the end-- that you do, do run that podcast and, and help with the education. That's fantastic.

Share This Episode:
By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.
Preferences
Deny
Accept