# Demystifying Decentralized Finance – Thomas Smets (Arcadia Finance) | Onchain Soirée [1]

- Channel: [ETHBelgium](https://streameth.org/ethbelgium)
- Date: 2025-10-07
- Duration: 16:02
- Watch: https://streameth.org/watch/yt-AZcNdTf7L3M
- YouTube: https://www.youtube.com/watch?v=AZcNdTf7L3M

## Description

Onchain Soirée [1] was hosted by ETHBelgium at KU Leuven on April 28, 2025; an intimate evening of lightning talks, meaningful conversations, and high-signal energy with researchers, founders, students, open source builders, lawyers, and the Web3-curious.

ETHBelgium is building Belgium’s most vibrant community of Web3 builders — one conversation, event, and hack at a time.

💬 Join the conversation: https://t.me/ethbelgium

🧠 Follow us for more:
• Twitter: https://twitter.com/ethbelgiumHQ
• LinkedIn: https://www.linkedin.com/company/ethbelgium

## Transcript

seems to be working. Thank you. Uh so quick words, who am I? Uh Thomas. I studied here uh so spent some time in these and uh since my time at university, I was already playing around with blockchain. Um mostly losing money trading, but I also delved into the tech side. So started building some smaller projects. Then I decided to join intellect in Belgium. So it's a company that builds more proof of concepts for traditional financial institutions but on the side I kept very deep in uh the public space. Started doing me um we like to invent new names for existing things. So it's basically arbitrageia but then uh with blockchain protocols and then since 2022 start building Arcadia finance. Um it's a D5 pro protocol where we focus on management of liquidity positions of decentralized exchanges. So very niches sector but I would say one with impact with product market fit. So we help market makers onchain to manage their assets. Uh but more on that later. Uh maybe a question to see the audience who knows what flesh loans are put your hands up. Okay, then I underestimated you. Um, so since I will start with some basic concepts to set the scene, DeFi what it is all about. If we go back way way way uh in ancient times, two people want to trade. Um, trading goes back 150,000 years in the past. How did it work? They meet up in a local market and they bring their goods. Human evolves. So of course what we trade expands both in quality and in distance. So after a while it becomes impractical to transport your goods to the place to market. So around 3000 before Christ, some smart guy saw some clay and some sticks on the ground in Mesopotamia and he thought, why don't we just write our trading deal on this clay tablet and we go to the market with that and we don't have to bring all our goods. So that was actually quite a good idea because now at the local marketplace, people just had to show up with a clay tablet to do trade. By the way, the first written records ever found by humankind are financial ledgers to store trading transactions. So, it's maybe a moment to reflect what we've building here is one of the most important technologies ever invented by humankind. Scripture, we can transfer information, knowledge over generations. And the reason we invented it was to have ledgers to trade. But now a new problem emerges. Um, under normal circumstances, our ledgers agree with each other. But what if we have now people that made a deal and when it has to be settled a few months later, they have different ledgers. Some clever guy saw an opportunity and decided to be the trusted middleman. So instead that each party keeps their own ledger, we have one party that keeps the ledger for a whole country, a whole community and this guy can tell every time who owns what. Of course technology uh improves. We invent scripture uh in the 1300s um dubal entry uh accounting later on we have the digital age. So the amount of uh information we can process increases. Um like now we have millions of transactions we can handle per seconds all over the globe. But the same principle remains the same. There is a single guy keeping the ledger for everyone storing all our trade transactions. Think you see where we are going until 2008. Now, for the first time, Satoshi Nakamoto, he finds a way um that we can do trades, that we all have our own ledger, but that we keep it in sync without the middleman. But why should you as a user care? That's uh where we're going to dive into now. So, here I show you two uh front ends, two consumer applications to trade. on the left side a D5 protocol on the right side a traditional financial broker. Do you see the differences? No. And I think that's how it should be. As a user, you should not care if you're using DeFi or centralized finance. As a user, you want to use something because it's better, because it's cheaper, because it can process transactions faster, because it's more robust, because you have new features that are unlocked. So while DI and CFI should look the same, I think in the end defi should win just because it's better than CFI and not because we want to use a blockchain underneath. And I think there are some good reasons uh while that is the case. So for a user it looks the same but what happens under the hood is different. In DeFi, we use a blockchain as the final ledger that says who owns what, which uh yeah, contracts are open standing and the logic is implemented by smart contracts, immutable um decentralized code. CFI, we have a trusted uh institution that has their own backend to enforce the rules and their own database to store it. So, how do these differ? It's decentralized DeFi, meaning um if one of the nodes is out, still the whole network works. While in the centralized world, if the thing is down, you can't do anything with it. And the advantage of making your slides last minute is that you can incorporate the news. Today in Spain and in Portugal, people are queuing for all the uh bank accounts since they cannot pay for the taxis, which they need because the train doesn't work. So this decentralization is not just a selling point. Over a long enough time, every central system will fail at some point and things should fail, but you should have backup. Next to that, it's trustless. Um yeah, I think we everyone who was in blockchain know how important it is. As soon as you have some middleman, at some point it will go wrong. It is permissionless. Uh if I want to build a blockchain game and I want to use unis swap to trade my uh rune schemters, unis swap cannot stop me. They built their code to work with any c20. No matter who builds on top of that can use it. Compare that with the app store. If I built a game on Apple, but I want to sell goods, either I can't or I pay 30% uh to our good friends. Also important here to note most applications don't involve a single um piece of logic a simple component but it consists of multiple while in the centralized world these are different ledgers that have to be synced with the blockchain network there is one blockchain where the different applications the different smart contracts are built on top and that brings me what I think is the most important difference that is atomic execution. Unfortunately, it's also the hardest to explain. Um, but let's use as example an exchange. In reality, unis swap, most known decentralized exchange. That's not one smart contracts. It is often 10 different smart contracts cooperation. You go through multiple pools. You have a router, a solver. Similar in the centralized world, if you trade stocks, there's not one single central party involved. You have your broker, you have your clearing house, you have your exchange, you have your central security deposit. All of these have their own ledgers. And when you want to do a trade, it first has to be cleared. So all of these intermediates have to agree that what you can do can be done and then it has to be settled. So they all sequentially do the actual transfers, change the entries in their database basically. But that doesn't happen at the same time in blockchain. However, here clearing and settlement happens at the same exact instance. So even though my trade goes to 1,000 different smart contracts, if at the end um one of them fails, all of them fail. And this is not a let's say imaginary risk. Uh I would say Google hex that bank and you will see what the risk is of having difference between clearing and settling trades. It's basically why it takes so long to send funds to America, why it's so expensive, etc. So we have decentralized crystless atomic composible. uh this should give us brilliant applications but where are they? So where are the actual decentralized applications? I have two theories. Um the first one is DeFi is built by nerds for nerds and I think we in auditorium computer science where do we nerds like to focus on? It is backends, smart contracts, databases, blockchain. Where do we not like to focus on front ends, user applications? So we have automaticity that we can simplify so much for the user, but we don't build it. We build very complicated, clutchy things. On the other hand, it is very easy to bash on the nerd. Uh I don't think this is completely true. I think we nerds need to stand up and return the favor like the traditional hype cycle. A lot is expected in the beginning. After that grows, compare this graph now with the graph DVLDI. I would say it's almost a perfect match. Defi only started 2020. That's 5 years ago. It's not fair to expect that we should have already replaced all banks for all things. We're still experimenting. We passed a big hype cycle, but we also achieved a lot of great things. For instance, um yeah, unfortunately the main uh product market fit is still trading and we see that decentralized exchanges are picking up month after month. more volume goes to the decentralized exchanges uh relative share and this is something I uh like got from the internet yesterday. You can verify it yourself. Us as a DeFi company, we have to do a lot of work together with both European and US clients. So I very often have to exchange dollars for euros or visa versa. The cheapest way to do that is on chain. So, here is $1,000 I want to exchange for euros. Wise has a nice uh application where they compare all their competitors. If I do it on base with Odos, you see I get $1,35 back uh euros back. The cheapest on thread is,130. So, I would say here we already beat the average uh no not the average the best centralized bank and I think a reason is because of atomicity. ODOS they are a um aggregator. So what they do is they look for the cheapest part to go from euros to dollars and they don't have to go through a single exchange. They can take into account slippage and fees of different exchanges and basically split up the whole flow over 15 different exchanges. Due to atomicity, this whole thing works or does not work. you can never end up that your transaction only goes to rep eat intermediately and not to USDC. So I would say a first use case that's already live today can be used I use it daily not daily but monthly. Secondly, there are already enduser applications, some of them that people don't even realize there is blockchain used underneath. I showed you Hyperlquid. This looks like a real exchange similar to Interactive Brokers, Dejiro, whatever you want. It has the same look and feel, the same speed. Only your assets are self-custodied. I think poly markets uh was quoted in most of the uh US news around the elections to see how uh the polls are let's say upto-date non-manipulated uses prediction markets underneath and then yeah maybe less DeFi but also social media is seeing their own um yeah their own uh applications and lastly actual finance is coming um there are a lot of the big institutions now thinking about tokenizing. Stable coins are having their product market fit. I think bonds will come on chain soon and uh Coinbase is now busy tokenizing stocks. Uh also Black Rockck is looking into it. So in five five year times that's not bad. I think we should be not too hard on ourselves. Of course I will keep it very quick because I'm already over time. I apologize. So what what does Arcadia play a role in here? uh we really believe that we should build anti- consumer applications. I think our target audience are market makers. So it's quite a niche sector. Um but for them we want to build actual uh usable applications. Uh so what is a market maker? Basically when we do this trade as a user you want to at this moment exchange dollar for euro but someone has to take the opposite of the trade. Same for sellers. A market maker facilitates trades in both directions. They post liquidity in the dexes and they charge a small fee either side and it is their profit and uh today that is very hard to do. For instance, for uh as a market maker building a position today, they have to do 19 steps, nine onchains transaction and that's just to build the position. After that, they have to manage it, hedge it, rebalance it, compound. With Arcadia, they can do that in a single transaction, see it over different dexes and soon different chains in the same application and they can automate um things like rebalancing, compounding and when I say we, I mean our smart contracts. We as a company take zero custody over any of these liquidity positions. Um yeah, so in the end, why do they care? It proves their user experience. it uh makes them earn more money um and they love it. This is it. Thank you. Some thanks Thomas. Absolutely awesome. Uh I think everyone is going to take some D5 with them uh today. Next up is uh Rafael.