Interview: Julien Bouteloup: Decentralized finance disruption?

After completing French preparatory classes, I studied computer science in the United States, specializing in machine learning. I then worked for the European Security and Defense. On a personal level, I took advantage of the data storage capacity and computing power I had at my disposal to run algorithms. The first one was doing astronomical analysis. The second was Bitcoin, in 2010.

Some time later, in 2015, I discovered Ethereum, which offered a whole new perspective. Where Bitcoin would be a monetary reserve, Ethereum would be the new world of the Internet. This foreshadowed a complete transformation of our society, and this is what immediately intrigued me and drove me to launch several projects.

Following the emergence of the first networks secured by the Proof of Stake[1], I created a staking-as-a-service platform, allowing diverse actors to validate the transactions of different networks via the pledging of their assets. This is how Stake Capital[2] was born.

Since then, I have created and participated in many decentralized projects, such as Stake DAO[3], Curve[4] [5] or BlackPool[6] [7].

[1] Proof of Stake - Wikipedia

[2] Stake Capital - Official website

[3] Stake DAO - Official website

[4] Curve - Official website

[5] DeFi Deep Dive #8: Curve Finance

[6] BlackPool - Official website

[7] BlackPool: DeFine by Stake DAO Ep 6

BlackPool was created a year ago as a result of a personal investment in the Sorare[8] [9] platform. While I started by playing myself, it soon became more interesting to employ experienced players in order to manage my assets more profitably.

BlackPool is the next step, it is a decentralized fund that collects NFTs from games, and delegates them to experienced players. The players get their compensation, and the fund collects the rest, allowing owners of BPT tokens, our governance token, to generate income.

At the moment, we are mainly focusing on play-to-earn NFTs, since they generate passive income. On the contrary, speculative NFTs like CryptoPunks[10] [11] are riskier and uncertain investments as they are mainly speculative and depend on market dynamics.

Our fund is divided into several "verticals[12]", one for each game we invest in, but also verticals of collectible NFTs, such as Hashmasks[13].

As an illustration, our Axie Infinity[14] vertical is populated by 634 scholars, who are people committed to playing with the NFTs we delegate them with. Most of them are young players from developing countries, and this mechanism allows them to earn a substantial income in respect to their economic zones, which have been particularly affected by the recent sanitary crisis[15].

[8] Sorare - Official website

[9] Sorare, the French unicorn built on fantasy football

[10] CryptoPunks - Official website

[11] The Fat CryptoPunks Thesis

[12] BlackPool verticals

[13] Hashmasks - Official website

[14] Axie Infinity - Official website

[15] Documentary: Axie Infinity and its impact on the Philippines

Stake Capital is the first entity we created, it is a staking-as-a-service platform allowing the pledging of digital assets with the aim of securing protocols and, in the process, earning a return on these assets.

Beyond the income generated, there is also a lobbying dimension due to the accumulation of voting rights by the network validators[16]. These voting rights indirectly allow the validators to intervene on the markets, and to reorganize the transactions to be validated in order to make a profit: this is known as MEV[17] [18] (Miner Extractable Value). Note that since January 1st of 2020, more than $700 million has been extracted by miners and other entities participating in MEV operations, on the Ethereum blockchain alone[19].

Today, we focus on other financial operations, such as liquidations[20], arbitrage[21], and market making[22]. We also have an investment-oriented dimension, just like any quantitative hedge fund, with a traditional approach, which operates mainly in the decentralized finance (DeFi) market. We have invested in more than a hundred projects.

Finally, we have recently moved into treasury management, offering institutional investors substantial returns from the DeFi infrastructure. This is exactly what we have done in the context of our partnership with SwissBorg[23], which can then remarket this return to its own clients.

[16] Validators are the actors who secure the network. Among their missions, they are in charge of validating all the transactions that pass through the network, and of creating blocks grouping these transactions.

[17] Miner Extractable Value (MEV)

[18] MEV Resources

[19] MEV Explore

[20] Liquidation definition

[21] Arbitrage - Wikipedia

[22] Market Maker definition

[23] Stake DAO x SuissBorg

MEV is the extraction of data from the pending transactions, and the exploitation of this data. When a user sends a transaction, let's imagine the transfer of an ether to a third party address, this transaction ends up in the mempool[24], which is an antechamber of pending transactions. Miners can then create blocks by collecting transactions from the mempool. The MEV occurs right before the validation by the miner, at the mempool level. It involves groups of actors who extract information from pending transactions, and use this information to their advantage. For example, a user may detect a price difference on the same asset pair in two different markets. The user will then perform an arbitrage transaction to balance the prices between these two markets and thus make a profit. Before being validated by a miner, this transaction will be in the mempool and can be seen by everyone, since the mempool is public. An actor will therefore be able to understand that this transaction is a risk-free arbitrage and will be able to copy this same transaction by offering a higher remuneration to the miner, and thus obtain a profit at the expense of the first user.

With Stake Capital and Stake DAO, we devote a significant budget to AWS (Amazon Web Services, Amazon's cloud service) in order to host our own validator nodes, and thus privilege our own MEV transactions. We also use EDEN Network[25] [26] to send our transactions directly to the miners without going through the mempool.

Unfortunately, the consequences of MEV are multiple for the end user: sandwich attacks[27], front-running[28], etc. Finally, the escalation of transaction fees by the MEV players leads to higher transaction fees for the average user.

[24] Mempool definition

[25] EDEN Network - Official website

[26] Multicoin Capital: Tokenizing MEV

[27] Sandwich Attacks Explained

[28] Front-running - Wikipedia

Several solutions have already been conceptualized and implemented to reduce the impact of MEV on the users.

Among them, Flashbots[29] allows to move the auctioning of MEV transactions outside the Ethereum blockchain, and to deport it to direct channels between miners and traders. The immediate impact is the decongestion of the Ethereum network, making transaction fees cheaper than usual.

I also mentioned EDEN Network, which creates a circular incentive structure between miners, traders and Slot Tenants[30]. Traders can send their trades directly to the network without going through the mempool, and they are processed in a way that maximizes the outcome for the entire ecosystem.

A third solution has been explored, and has even been implemented by Optimism[31], one of Ethereum's scaling solutions. This solution consists of designating an actor who will be responsible for sequencing transactions for a given period. The designation of this actor is done via an auction, which is why this concept is called MEVA: MEV Auctions[32]. This can obviously lead to threats as an actor could corrupt the system via his auction and prioritize the transactions in a malicious way, but combining this idea with randomization and a reputation system could be interesting to explore. In any case, it is always a question of reconciling the economic incentives of all actors to achieve a circular system that benefits the whole.

[29] Flashbots: Frontrunning the MEV Crisis

[30] EDEN Network: Slot Tenants

[31] Paradigm Research: How does Optimism's Rollup really work?

[32] MEV Auction: Auctioning transaction ordering rights as a solution to Miner Extractable Value

First, it would be convenient to explain the concept of flash loan. It was introduced by Aave[33], a platform that allows users to contract a loan against a collateral deposit, just like credit lines in traditional finance. The flash loan is a debt where the collateral deposit is not necessary, since the debt is contracted and then repaid in the same transaction. The risk of default is therefore zero (the transaction, and therefore the loan, can only be executed if the loan is repaid in the same transaction as the loan). This concept thus allows one to request liquidity that is not necessarily possessed, and to take actions on the market with this liquidity.

When this feature was released, I contacted the Aave team in order to validate the mechanism, which allowed me to really experiment with it. There were few opportunities at the time and the cost of borrowing was relatively high, so it was mostly done for fun.

This concept facilitates fundamental market mechanisms, such as liquidations or arbitrage, which otherwise the market would not function. It also allows anyone to perform actions that only venture capital funds or hedge funds could previously undertake.

[33] Aave overview

I started writing articles in 2017, in which I exposed the frauds and scams coming from ICOs. As a result, I was sued several times for defamatory statements, for sharing information that was proven to be true. I then realized that I had to change my approach, and I created a new media with two partners. We were in complete agreement on the editorial line since we wanted to create an "underground" investigative technology media focused on hacks in the DeFi space.

It should be noted that Rekt aims to be an agnostic media without bias, and that the reader is encouraged to inform himself and to form his own opinion on a given subject.

We have also incorporated some hidden elements into our project, namely the parlor system[34] that is not directly accessible from our website, or the physical journal that we published on the occasion of the 2021 Ethereum Community Conference in Paris.

[34] Rekt - Parlor

My team and I have shared this vision for some time. The production was done by a person who contacted me almost a year ago and whom I have integrated into my team.

To summarize this vision, it refers to the conflict led by nation-states in order to destroy or control certain protocols, before a total decentralization occurs. Hopium Diaries allowed us to formulate this vision, and to make it more tangible.

To protect themselves from hacks, developers must first ensure that their code is robust and flawless. It is essential to obtain an external opinion on its quality by auditing it.

Beyond the quality of the code and the vulnerabilities it may contain, it is necessary to think ahead and anticipate the uses that could be made of it. Considering the general use case of the program only can be dangerous because it is generally not isolated. Other protocols revolve around this program, and allow interaction with it. It is therefore necessary to take into account all the economic or governance attacks that it could suffer, including the use of flash loans.

To prevent this, the first thing to do is to isolate the components of the smart contract. For example, when comparing the architecture of Balancer[35] and Curve, we realize the difference in terms of resilience between these two protocols. Balancer uses the same Context for all its liquidity pools, which means that a siphon on one of the pools implies that the others may also be subject to this siphon. Curve, on the other hand, isolates each of its reserves, which limits the attack surface if one of them were to be affected.

It may also be worth considering whether the program can be isolated from the rest of the ecosystem. The disadvantage is that it cannot be integrated with other protocols, but it will be more secure.

Finally, it is essential to take into consideration the external opinion of a tokenomics specialist, who will be able to identify possible sources of economic or governance attacks.

On the other hand, as a preventive measure, it is essential to insure the protocol. This can be done via decentralized insurance protocols, such as Nexus Mutual[36]. This is what we have done with Stake DAO.

However, the ideal would be to subscribe to a traditional insurance contract, especially since it would come from a system that does not have the same economic environment. It would be interesting if decentralized insurances like Nexus Mutual were themselves insured by traditional insurances, in the manner of reinsurers like Scor[37].

[35] Fernando Martinelli, founder and CEO of Balancer: "My Vision for Balancer Protocol"

[36] Nexus Mutual and Defensibility in Cryptonative Insurance

[37] Scor - Wikipedia

Stake DAO's primary objective is to enter the DEX (decentralized exchange) market. We have known about it for some time, but CEXs (centralized exchanges) such as Binance, Coinbase or FTX will face strong regulatory constraints. This is already the case for Binance, which is under scrutiny by several jurisdictions[38]. Over time, we believe that the centralized exchange market will shift to decentralized exchange systems. This is a significant market, concentrating large amounts of capital and financial streams, and Stake DAO's goal is to acquire a share of it.

To do this, we need to offer a user experience similar to CEXs', offer attractive prices, but also put in place complementary services that allow one to earn on his assets. We have therefore created investment strategies for Euro, Dollar, Bitcoin and Ether with competitive returns. In addition, we also want to make derivatives available to users, such as options, futures, etc.

The goal is certainly not to provide access to the entire world of DeFi, as innovation is far too fast for anyone to master it in its entirety, but rather to provide access to a carefully selected list of products so that our users can make the most out of this market.

When we launched Stake DAO, we first implemented investment strategies because they allow us to accumulate voting rights on Curve. Curve is a DEX that allows its users to trade tokens with very little price slippage[39]. This no price slippage trading is allowed because the liquidity pairs available on Curve all contain assets that are similarly priced to each other. Accumulating these voting rights then allows us to leverage Curve, improve the performance of our investment strategies, which are primarily based on this DEX, and allow our users to trade their tokens at lower cost. It's important to note that to be whitelisted by Curve, we had to invest nearly $45 million, at a time when their governance token was much less valued than it is today. There is a form of power play around this protocol, where several entities are trying to capture as many voting rights as possible[40]. Today, Stake DAO is on their whitelist[41], and we are helping other decentralized protocols like Frax[42], and soon Rari Capital[43], to leverage Curve.

[38] Binance and regulatory scrutiny: changing times for the crypto market

[39] Slippage definition

[40] Rekt - Curve Wars

[41] Stake DAO x Curve

[42] Alpha Leak: The Algo Coin to End All Algo Coins - Sam Kazemian, Frax

[43] Rari Capital: How to borrow and lend any token

Indeed, in order for Curve to mitigate price slippage, and to offer users the ability to trade large amounts without consequence, it was initially chosen to only set up pairs of assets with similar prices. For example, it is possible to trade USDT against USDC, both of which are intended to be indexed to the dollar. It is also possible to trade WBTC for sBTC, which are both indexed to Bitcoin. It is therefore not possible to unbalance the liquidity pools[44] in terms of valuation, since all the assets in each pool have a similar price. Consequently, liquidity providers cannot suffer impermanent loss[45], and users cannot suffer price slippage.

Only, mathematics evolves, and the tricrypto pool is the result of this evolution[46]. The simulations of this liquidity pool show us that price slippage and impermanent loss are strongly mitigated, despite the difference in prices of the assets that make up the tricrypto pool.

[44] How do liquidity pools work?

[45] What is impermanent loss?

[46] Curve v2 whitepaper

Several methods can be implemented to avoid the impermanent loss on exchanges between assets having dissimilar prices.

The one we have implemented on Curve is in collaboration with Synthetix[47]. The idea is simple: Synthetix allows users to create synthetic assets using their governance token. Users deposit SNX into the protocol, and can borrow against that deposit at a 750% collateralization ratio. This debt takes the form of the chosen synthetic asset, which will be indexed to the underlying asset. One can therefore deposit SNX and borrow sBTC, sETH or sUSD which will follow the price of their underlying asset (i.e. BTC, ETH or USD). This mechanism enables, for example, to exchange sBTC for sUSD without any price movement since these assets are only a virtual representation of the underlying asset, and they draw their value from the collateral: SNX. So let's imagine that we want to exchange BTC for DAI[48] [49]. To do this, we exchange BTC for sBTC via a Curve pool, the sBTC is then exchanged for sUSD on Synthetix without any price slippage, and then the sUSD is converted to DAI on Curve. This results in a transaction with very little price slippage, and the liquidity providers also have very little exposure to impermanent loss.

[47] Synthetix - Official website

[48] Stablecoin pegged to the dollar.

[49] Dai (then called the "eDollar") was described in a reddit post