🎙 "We Said Let's Start With Uniswap and See What We Can Unbundle:" Pranay Mohan of Hashflow
In this week’s episode, we talk with Varun Kumar and Pranay Mohan from Hashflow, a new decentralized exchange that wants to unbundle Uniswap. What this means is, they want to keep the non-custodial side and the liquidity pools, but instead of finding price...
By:
In this week’s episode, we talk with Varun Kumar and Pranay Mohan from Hashflow, a new decentralized exchange that wants to unbundle Uniswap. What this means is, they want to keep the non-custodial side and the liquidity pools, but instead of finding prices with smart contacts and a formula, they use market makers applying their highly specialized strategies and algorithms behind the scenes. Anyone can become a market maker, and some may open their pools so that others will be able to share their capital in exchange for a portion of trading fees. Their bet is that they will combine machines and humans in just the right way, to provide cheaper rates at lower prices.
We talk about their launch, their upcoming token, and more broadly, we dive into their big vision for Hashflow and DeFi —and it's that one day all of the world's liquidity will flow through open networks.
They explain why they didn’t build on Layer 2 -- and it seems like these founders are cautious enough about the state of scaling solutions that they decided to hold off until they’re fully convinced.
We also got philosophical, talking about the history of money and how crypto allows us to break away from a debt-based system that’s forced upon us to a voluntary, incentives-based design where each community can choose and experiment with monetary theory and policies, and manifest that into code.
But first we get into how these aerospace and chemical engineers got into building a decentralized exchange.
The podcast was led by Camila Russo, and edited by Alp Gasimov. Transcript was edited by Owen Fernau and Dan Kahan.
🎙Listen to the interview in this week’s podcast episode here:
You’re a paid subscriber, which means you get the full transcript below. Subscribers also get exclusive access to The Defiant’s Discord chat for the community, here’s a new link to join.
🙌 Together with:
- Zerion, a simple interface to access and use decentralized finance
- Balancer, one of the leading DeFi automated market makers (AMM) for multiple tokens. Dive into their pools at https://balancer.finance/!
- Kraken, consistently rated the best and most secure cryptocurrency exchange, which can get you from fiat to DeFi
- Aave, an open source and non-custodial liquidity protocol where users can earn interest on deposits and borrow assets.
Varun Kumar: Absolutely. Thank you so much for the kind introduction and I'm super excited to be here. So yes, I am the founder and chief entertainment officer at Hashflow, so for entertainment value. So that's it. Yeah, the story is quite interesting. So before crypto, obviously, I was an engineering student studying my Ph.D. at Stanford, and ended up never doing my Ph.D., and joining the industry and becoming a big boy. And then in the interim, the course, I think I met a friend called Ric Burton in a hot tub, and that set the premise for my journey into crypto.
Camila Russo: That sets the tone pretty well for crypto.
VK: Yes. So it was one of those things in San Francisco, we’re sitting there and Ric said, you should look into Ethereum. Like, you know, if you're building a satellite system, I'm pretty sure you can figure out Solidity. And I'm like, that sounds interesting, but where do I begin, right? And that's how the discussion started, and then it went into the whole notion of money and such. And then that took me down the rabbit hole, just the discovery of how money works.
My first realization was, I had no idea how money works. I'm like, okay, money is something that grows in the farm, I reckon, but turns out that's not the case. So that took me down a rabbit hole, which we can talk about after Pranay’s introduction. It’s an interesting story of how the dots connect and how one thing led to another and that led to finally what we see on Hashflow today. So Pranay, go ahead. Sorry.
Pranay Mohan: Yeah, my background is not as entertaining. But my background is somewhat similar to Varun in that I also come from a hard engineering background. I studied chemical engineering and started my career in oil and gas and CPG, or Consumer Packaged Goods. And I think the inflection point for me to start migrating into tech and software was in 2013 when the Snowden revelations came out, I was working on a project at Frito-Lay to make the best potato chip. And I'm not kidding, that was literally my project to make potato chip slices more standardized so that at the margins, Frito-Lay could save millions of dollars.
And so when the Snowden leaks happened, the ennui of what I was experiencing as a potato chip engineer bubbled up to the surface. What Snowden shined the light on in terms of the surveillance dragnet that all of us were a part of made me realize that the frontier no longer existed in the physical world, that it had moved to the digital realm. And if I was going to play any role in dealing with the human rights challenges that will emerge in the next 100 years, I need to move to software, I need to become a technologist. And so that catalyzed that movement into tech. But like every kind of aspiring young techie, I ended up at big tech companies putting dog filters on people.
“What Snowden shined the light on in terms of the surveillance dragnet that all of us were a part of made me realize that the frontier no longer existed in the physical world, that it had moved to the digital realm. And if I was going to play any role in dealing with the human rights challenges that will emerge in the next 100 years, I need to move to software, I need to become a technologist.”
So in 2017, when the whole crypto boom happened, I started reading up on it, getting more interested. And in 2018, I helped a friend host a conference where I met some really brilliant people in the space. So people like Olaf Carlson-Wee from Polychain, Jutta Steiner from Parity, Izzy Meckler from Mina. And I realized, okay, this isn't just a virtual currency trading, it's not just speculation, there are these really brilliant people building some really cool shit in this space. And so I finally felt like I had found my “Rebel Alliance”, and that's what put me into that proverbial rabbit hole. And now it's a mind virus, I can't think about anything else. So that's how I'm here chatting with you.
Ideological Alignment
CR: Nice. So how did you two meet?
VK: We met at Stanford Blockchain Conference last year discussing philosophy of money. I was telling Pranay about the journey of why I got into crypto. I think TL;DR was the realization that essentially, money in its current form, what we called fiat money, is a modern form of slavery. And moving away from that, moving away from a debt-based monetary system to a user-based incentive design choice was a powerful narrative for me to look into crypto in general.
To take a few steps back, if you go all the way to bartering, and think of that as a giant mesh network of people trading items with each other, that’s the most primitive fundamental form of trading, where humans are trading items with each other. And they have some value attached to it, but really, it's just goods changing hands. And then eventually, what happened is some commodities became more prominent or traded more often than the others.
In fact, we started seeing a pattern where some commodities became the standard through which all trades started taking place. So we could say that the technology of money was discovered, where there was an incentive for most people to own certain items and use them to trade other goods. So you can see this transition from a mesh network to a multi-centric, but distributed system, where some commodities became the money.
“With crypto, what is interesting to me is we actually get to break that will, where we get to get away from a debt-based system that is essentially forced upon you through the logical violence to two-way system of user-based incentive design where you get to use the money of your choice. You can freely experiment with monetary theory and policies, and manifest that into code, and anybody can have the choice to use the money they want to use, which is what inspired me to get into crypto.”
But then we realized, well, this is great. But if you want to scale this to geospatial regions across continents to conduct trade, scaling physical money is kind of hard, right? So the way we solved it, the way we overcame the physical constraints of commodity money, was by inventing something called currency, which we can think of as a Layer 2 scaling solution to the underlying hard money whereby you had these custodians emerge who would keep your money in custody, and then issue claim checks that could be used to redeem the underlying money. And this paper could be used easily to scale across different regions, but institutions would recognize these bills, and they will give you the money that's sitting there.
This could be seen as without the first fundamental modern form of banking system where institutions would issue you claim checks that you could use to scale and then they will keep money in custody, which eventually led to the John Law system later. But then even before, fast forward 1600s to 1900s, what we saw as the gold standard until 1913 when the Federal Reserve became a thing.
The emergence of Federal Reserve, I almost see that as a forced network update on the system to move away from Gold Standard to what we today call fiat. With gold standard, we had currency backed by money where money was gold, and currency was essentially a dollar bill, a claim check, an underlying money. But the moment you move away from that, you change or you get rid of the money paid and currency paid and merged into one. That's called fiat, which is essentially what we had after Bretton Woods, after the 60s, what we use today.
By fundamental design, it has to be borrowed into existence, and then you as a network user of the US network are forced to pay rent for its usage in the form of taxes and fees. But the only way for you to pay that rent in the same currency that you borrow is to borrow more. So the only way you could do that is by borrowing more from the system. As a result, it's a perpetual debt-based cycle or infinite regression, but there's no point of time where the user would be truly debt free. Really the only person who wins in the system is the one who controls the system, right, so it's a zero-sum game.
And with crypto, what is interesting to me is we actually get to break that will, where we get to get away from a debt-based system that is essentially forced upon you through the logical violence to two-way system of user-based incentive design where you get to use the money of your choice. You can freely experiment with monetary theory and policies, and manifest that into code, and anybody can have the choice to use the money they want to use, which is what inspired me to get into crypto. That was the discussion we were having. And then we were like, okay, I think we agree on many, many things, we should probably work together. So I think I spoke too much, but Pranay, you can add something there.
PM: Yeah. I mean, I think the general vibe is we both geek out about what is the fabric of the monetary system that we live in? I think at the same time, I had been reading some of Ray Dalio, his work. And also, I follow an author, a financial adviser online, Lyn Alden, who has written some really great stuff around ‘where is the dollar now and why are we in this part of the cycle we're in.’ Tying into what Varun is saying, like the petro dollar, the switch from Bretton Woods, the gold standard, over to what is the petro dollar system now is not only affecting the rest of the world, it's not just putting other folks in bonding to a world reserve asset, it also affects the US consumer. Because the US gets its own manufacturing economy in order to run a constant deficit in order to put the dollar out into the world to service all the debt.
So having an opt out, having an alternative to the system gives you a lot of hope. It takes something that feels irredeemable and gives you a pathway out. And so when you meet like-minded folks who are eager and understand the bigger picture of what we're all trying to do with all this, that can look past the rugs and the scams and the hot dog tokens, it really gives you an anchor to the fact that you're working towards a 10, 20, 50, 100 year future. And you want to find people like that and work with them because you know they are anchoring to building a world that gives people more rights and more freedoms from the system that they have now.
Future Financial Systems
CR: I love this big picture on why crypto is important and revolutionary and why the old system is broken. To you guys, what's the end goal or the end state if everything continues progressing as it has and crypto ends up taking over finance. Do we have a single global currency that's maybe Bitcoin or Ether, or is it like multiple currencies based on where people are using these networks? How do you see this playing out if crypto were actually able to replace the current financial system?
VK: That's a good question. The way I see it is crypto is to money what the internet was to communication standards back in the 90s. So previously, something was a monopoly, and then that monopoly was disrupted, and now anybody can freely deploy, it’s a free market. So that has been applied to the idea of money, which happens to be the biggest monopoly out there.
So as a result, what's interesting with crypto is many of the problems you see with fiat, like deflationary currency, inflationary currency, or even more philosophical discussions like capitalism or socialism and such constructs, they’re obsolete to much degree. They become irrelevant the moment you say the boundary conditions for the money plane is removed. Only if you constrain everybody to a single monetary system, you have those issues. But the moment you're no longer under constraint to play within the rules of one system, that could be the end of different systems, and those issues become almost obsolete or irrelevant. Now you're operating in a very orthogonal frame of reference. Quite honestly, we haven't seen anything like that before. So it's hard to say.
At the same time, because there is no more one currency to rule them all system anymore, we don't need to have this notion that we need a global currency. I think the only reason we think that way is because we've been conditioned to think that the dollar is the most powerful currency and we need to think of everything in terms of dollars. But the moment you see that that boundary condition has been lifted, then your money is essentially a commodity just like your phone. You can have an iPhone, you can have a Samsung, you can have a local phone in China, Huawei, or Samsung India, so you could do the same thing with money. You could have several nation states with their own currencies, you could have a single currency, we do not know. Right? But the good thing is it's a choice. It's no longer a monopoly, which was true of fiat money.
“Only if you constrain everybody to a single monetary system, you have those issues. But the moment you're no longer under constraint to play within the rules of one system, that could be the end of different systems, and those issues become almost obsolete or irrelevant.”
PM: Yeah, I think the concept of removing the boundary conditions is really important, because I think it would be naive to say that what we're building right now, Bitcoin is going to replace...I'm not a Bitcoin maximalist, it's not going to replace the dollar, and we're all going to just get on the Bitcoin standard.
But the important key is realizing that optionality is what we're missing, you're lacking choice in the current system, because for whatever reason it's evolved to be this way. I think Bucky from Cosmos talks about this a lot, removing the dependence on one world currency and having an ecology of currencies that people can use and play with. And so I think what we are in right now is in the very early stages of exploring that whole new design space that has been guarded off, as Varun was saying.
There's a really wonderful article online called “Entering the Crypto Idea maze” that talks about the three phases of crypto exploration, it starts with sound money, because that's the rise stone, the first thing people find when they engage with the concept of money. Then they move to the concept of open finance or DeFi, which is beyond the money primitive, beyond just currency, what can we do with this, what other financial services can we provide? And then once you have that framework, those primitives in place, then you can move to Web 3.0, the truly open web where you're using these financial primitives that you built to create Information commons that has value baked in.
If you tie back to what people were saying at Bretton Woods, Robert Triffin was saying, this isn't going to work, like it's going to fail. We're seeing the beginning of that now, we're seeing the fraying of that dollar base system that has made all of us prosperous. But I think we're seeing the winding down of the old story, and we're entering a new one. And so I couldn't feel more grateful to be in a position to be a part of this and play a role in shaping what that next 100 year future looks like.
Building for Web3
CR: Yeah, definitely it does seem revolutionary, the advent of this new Web 3.0 system, and it feels like COVID has played an accelerator role in this with central banks just printing even more fiat currency and trying to shore up their slowing economies with stimulus. That all just seems to highlight how fragile the old system is. And just makes clear why open finance and crypto should be taking over. So how did you guys get from these ideas into making...what inspired Hashflow?
VK: That's a really good question. Last year we were tinkering around with the idea of just regular L2s, should we build a Layer 2 scaling system to scale finance or what do we solve? Because clearly, Uniswap was a winner, I think Pranay has a good insight into that. Pranay, you want to take that question?
PM: Yeah, so we had DeFi summer, right. Cami, I watched your interview with Jim Bianco, and it's like super cool to see somebody from the TradFi world that is so psyched and being essentially a hype man for DeFi. And a lot of that was possible thanks to DeFi summer, the launch of Uniswap v2, and it catalyzing the basic economic primitive in DeFi which is being able to trade assets. Without that Lego block, you're not able to build anything on top of that. So that Lego block had to reveal itself in order for us to have this vibrant ecosystem that we have now.
So when we're thinking about what we could build, what is the set of possibilities of things here? One thing we realized is a lot of the motivations behind DeFi summer, the earliest adopters were ideologues. And I don't mean that in a bad way, like a lot of us that come to this space anchored to this idea of decentralization, that this is the core tenet and that's what matters. So when we explore the design space, we start off at the very end of the continuum of decentralization. We've seen what's gone wrong in a centralized world so we got to build the most absolute decentralized version of it. And that's what Uniswap essentially represents.
“But there's this whole middle space of design that is relatively underexplored, because as we were saying, we're just so early. So one thing that we were seeing is, maybe let's look at that Uniswap side, let's start at the Uniswap side of the decentralization continuum or the exchange continuum, and see what can we unbundle, what can we pick out of here and take from the more centralized side, and build it better for end-users?”
If you think about the design space of any type of exchange from decentralized to centralized, at one end, you have the perfectly, always on AMM that acts like an asset vending machine. It's a token robot where you go in, you put one token in and another one pops out, you don't have to talk to anybody. The vending machine pricing is a little fuzzy, but you're going to get a token out. On the other hand, you have something like Coinbase or Binance, where you have a centralized order book that runs on Coinbase’s server, you don't get self-custody, and all your transactions are intermediated by Coinbase, you don't have much insight into the settlement. And so that represents the other end of the pool.
But there's this whole middle space of design that is relatively underexplored, because as we were saying, we're just so early. So one thing that we were seeing is, maybe let's look at that Uniswap side, let's start at the Uniswap side of the decentralization continuum or the exchange continuum, and see what can we unbundle, what can we pick out of here and take from the more centralized side, and build it better for end-users? Software engineers have the same, there's no good or bad, there's only tradeoffs. So what is a tradeoff that Uniswap has made that may offer some other properties that users could benefit from?
And one of these things is the pricing. So one pattern we kept seeing is, every week there was a new curve, like Uniswap had a curve, and then Stable Swaps came out. And then you had these new AMMs that use oracles to improve the curve. But everything was just fidgeting with the curve, right? What if we just got rid of the curve? Why do we need that pricing to happen on-chain? Maybe you lose some quality of always on, but what if you can have that pricing happen off-chain, and then bridge it on-chain? You can potentially avoid doing that computation on-chain, you avoid anchoring to some programmatic pricing algorithm that you have to lock in a smart contract, and you can tap into the rich world of data that exists off-chain when you price assets. I think that was the key insight that led to Hashflow. I don't want to keep rambling, so I'll pass it to Varun.
VK: I think you're also interviewing a bunch of other market makers who traditionally market make on CeFi. Like why haven't you guys been doing DeFi market making? Regulatory concerns are one, but what else? I think one of the big things that kept coming up is we don't want to be constrained to X times Y equals K. So if you're spending tons of capital every day, or time, or compute hours to figure out the best pricing for an asset, then we don't want to be limited in what we could do in terms of how can we market make.
So really, what I was hearing is a really more granular control on how we price assets, and that could improve the general market structure. And from a user standpoint, they care about simplicity, they just want to go there and enter a number and get the vending machine experience. What we want to do is marry the whole price discovery aspect of market makers with the vending machine aspect of consumer demand.
And if you can bridge those two things, then we have Hashflow. If you're using Hashflow right now, since yesterday, you will notice that it very much feels like Uniswap, but from a pricing standpoint, or from a gas consumption standpoint, it's extremely efficient, and obviously, the way the quotes are coming from basically interfacing with people who've been perfecting the pricing strategy for the last 100 years.
Utilizing Market Makers
CR: So for listeners who might not be aware, this formula that Varun and Pranay were alluding to, X times Y equals K, that's the famous Uniswap formula for pricing. And so instead of using that formula and that pricing curve, what you guys do is have market makers come up with the best quotes for different tokens basically?
VK: Yeah, right. Because by doing so, you also are no longer requiring as much capital in the contracts. You also don't see as many of the other issues you would typically run into as a function of the curve. So obviously, the tradeoff like Pranay said, is it may not be on all the time. But on the other hand, you're able to get better pricing and you're able to deploy your capital in a much more efficient manner whether you are essentially using your smart contracts to deploy or maintain your inventory, and then quote against it, and sign that with your public private key combo. That way, you're always able to verify that the market maker was the one who signed the quote.
And from a user standpoint, it's extremely permissionless. Or maybe if I’m getting too technical, maybe Pranay, you could put on your professor hat and to make it a bit more audience friendly.
PM: Happy to, yeah. I think the metaphor that has worked best when I've talked to people or explained it is, it's like an OTC desk on chain. So changing what OTC means from meaning over the counter, which I don't think anybody actually goes to a counter anymore, to mean over the chain. So you can show up at the Hashflow app and just like Uniswap or Sushiswap, you type in what asset you want, and the volume of the asset you're swapping, and you get that vending machine experience, the number comes back.
“But what's happening behind the scenes is there's a lot of market makers with algorithms that bake into the quote, like historical asset prices, volatility, maybe sentiment analysis, things that people have been doing in the traditional finance space for so long.”
But what's happening under the hood is you're sending an RFQ or Request for Quote, to Hashflow which then farms it out to market makers, and then market makers respond back with a quote for the assets that you requested. And when that quote comes back to the user, they can choose whether to accept it, or just let it expire. If they accept it, then the signed quote is submitted on-chain to Hashflow smart contracts, and the transaction settles just like magic. So you keep the self-custodial aspect, you keep the decentralized settlement, and you have that very easy and simple user experience.
But what's happening behind the scenes is there's a lot of market makers with algorithms that bake into the quote, like historical asset prices, volatility, maybe sentiment analysis, things that people have been doing in the traditional finance space for so long. All of that is happening under the hood and giving you a very bespoke quote. And some of the benefits of this is that the price is much more granular, it's fine-tuned based on the conditions of the market that you can't observe on-chain.
And additionally, because you're settling this quote, and not having to compute that price on-chain, the gas is much, much lower as well. So in an L1 world where we're constantly dealing with $100 gas fees, if you're able to send a transaction with a $50 gas cost and you get the same swap experience, then why not?
Special Sauce
CR: And so, backstage, what are market makers doing? How do they come up with better quotes? Are they arbitraging with other DEXs or how does that happen?
VK: So typically, they don't disclose their strategies. That's their mode. In fact, many of the big hedge funds and market makers are basically, the way they make money for living is by having these proprietary algorithms that they use to get edge on others in the trading world. So as a result, allowing them to deploy those things to [inaudible 28:14], or something similar, makes for a much more competitive market, especially they can do it freely.
What's interesting is with the AMM model, you've got pools where any lazy RP can contribute capital and then your smart contract acts as the market maker, which takes money, and then it does market making and gives you a quote, gives you a price. So as a human, you're not really setting a price, you're just giving capital and then meeting us sitting back to be arbitraged by somebody else. And your contract will always give you a quote based on the inventory of assets it has.
It has its pros and cons. But then to focus more on the difference, in this case, what we're doing is instead of having anyone give you a capital, you're basically putting capital but you're also giving a quote on top. That's the primitive model. Now, what could be interesting although is if there's a world where market makers start creating public pools, where they bring their pricing strategy, but that people can contribute capital to the pool, it almost becomes where you let LPs meet best prices or best price makers, and it actually allows for a new way to scale things.
So then obviously, the thing is about how can you make the system more trustless, and that is something that we are announcing in the next iteration of Hashflow and how we are going to achieve that. But the fundamental idea is price discovery is usually proprietary, but that's also the secret sauce that leads to better pricing and better market structure.
“What could be interesting is if there's a world where market makers start creating public pools, where they bring their pricing strategy, but that people can contribute capital to the pool, it almost becomes where you let LPs meet best prices or best price makers, and it actually allows for a new way to scale things.”
CR: So you can still provide liquidity to Hashflow, that's still open, and you can earn LP fees from providing liquidity?
VK: Yeah, you can create your own pool, and you can start giving us quotes. So you can do that. And there are a few pools where you can actually also have other people contribute to your pool, and then you do market making, and then you basically split the rewards pro-rata based on equity ownership of the pool. So I don’t know if that makes sense, or Pranay you want to add something to that?
PM: Yeah. I think the key is that if you know the market maker, if the market maker has a good reputation and believes that they can give you a good yield, then it's possible for you to unbundle the capital allocation and the pricing. So Haseeb Qureshi of Dragonfly, who is a lead investor in Hashflow, had a really great article called “Unbundling Uniswap” where he breaks down what are the core components that make up Uniswap’s magic. And one of those is permissionless inventory, so allowing people, liquidity providers, retail LPs to go dump tokens in a liquidity pool, and now you have a virtualized counterparty. Like, you don't have one person providing the assets, and another person pricing; you have a crowdsourced inventory pool, and then the smart contract prices for you.
But if you do have a proprietary trading algorithm, you've spent a lot of money harvesting all this data, and you have the secret sauce, and you're getting great yield on it, capital is a commodity these days, you don't need to open up your pool to other people. But if you are an upstart market maker, it's like kaggle or something for data science, if you have some type of proprietary algorithm that you think will beat the rest of the competition but you don't know where to access capital, you can't find the liquidity, you can't open up your pool and get other people to take LP shares in your pool and say, hey, we're going to offer really great quotes and we're going to get yield, and I just need you to help me bootstrap this pool that we're starting together.
So anything is possible, but it really depends on what is the profile of the market maker that is spinning up the pool. Is it somebody who just wants to bring their pricing and they know where they can find capital? Or is it an upstart that needs a bootstrapping for that capital?
CR: Oh, interesting. So you may be opening a whole new space for DeFi innovation and yield, like people who want to become more specialized market makers?
VK: Yeah. So perhaps instead of just relying on yield from, let's say, a Sushiswap pool, now you can actually get some yield from Galaxy’s market making strategies, or Genesis’ market making strategies because you think they know how to price assets better than a smart contract.
CR: So anyone is able to contribute their capital to, say, Genesis market making for on Hashflow?
VK: As long as Genesis is okay with it. So if they are open to bootstrapping more capital, and they want to do it that way, then yes, they can do it. And looks like there is some interest from some of the market makers, who are building on Hashflow at the moment, so that could very well be what we anticipate next.
CR: Cool. And then otherwise, anyone is also free to create their own pool and strategy?
VK: Yeah, exactly.
CR: And so who are the market makers that you have right now?
VK: Are we allowed to say the names?
CR: Yes.
VK: Okay, let's put it this way. I'll tell you who invested in the company and I’ll let you deduce from that. So part of our fundraising strategies was to have people who we think are potential customers to also invest. So obviously, the round was led by Electric and Dragonfly, and then we had IDEO and Metastable and such. But then also on the market making side, we have Galaxy, and BCG, and Alemedaand LedgerPrime and BlockTower, Wintermute, etc. so there’re more names. So as a result, that gave us a pretty good initial bootstrap way to get to many of the PMMs which was a bit of a challenge early on.
But for the full disclosure then I think that for the first alpha round, right now, the pricing is coming from LedgerPrime, and a couple other market makers who for legal reasons, I cannot disclose the names just yet.
CR: Okay. Can you say how much you raised in this round?
VK: Yes. So we did a 3.2 million round. So it was fairly small, I reckon, if that is considered a small number. So that should be enough for us to get to the things we want to do.
Capital Efficiency Trumps TVL
CR: Nice. And so how much liquidity do you have right now?
VK: So, it's an interesting question. So in the alpha phase right now, we have about half a million in the contracts. But really, I don't think that's considered liquidity. Because the way I see it, I see liquidity is anything you can buy or sell. As an example, in the case of Airbnb, liquidity is not the number of people signing up for Airbnb; liquidity is how many listings are available. So same with Craigslist, liquidity is not necessarily the number of accounts, but also the number of listings for anything that's being advertised. So in an exchange, liquidity is not just capital sitting there, liquidity is how many bids in assets are sitting there that are ready to be taken or made.
“...liquidity is anything you can buy or sell. As an example, in the case of Airbnb, liquidity is not the number of people signing up for Airbnb; liquidity is how many listings are available. So same with Craigslist, liquidity is not necessarily the number of accounts, but also the number of listings for anything that's being advertised.”
So in Uniswap’s case, it's interesting it's called liquidity because you're contributing capital, but that capital is automatically being put to use by the contract in the form of quotes. So in this case, liquidity would be more so. Raw capital sitting in the contract is just an inventory. But the moment someone starts quoting on that, it becomes liquidity. So I would say right now, the inventory is somewhere near half a million, but the good thing is we can put 100% of that to use and the market makers can use that to quote and hedge their risk accordingly. So the liquidity is much, much greater than that number.
PM: Yeah, I think what the challenge is that TVL has become a meme number at this point. Like, Bitcoin has number go up in the price and then DeFi needed a number go up, so everybody gravitated around TVL, so we can watch that DeFi Pulse chart just up and to the right. But TVL is an important metric to understand how much raw potential there is in the system. But using TVL to compare Uniswap, Maker, Compound, Lightning, all of these different DeFi projects, it's kind of nonsensical, because that raw amount of assets available gets applied differently in each system.
“Bitcoin has number go up in the price and then DeFi needed a number go up, so everybody gravitated around TVL, so we can watch that DeFi Pulse chart just up and to the right.”
But what we're noticing is that as DeFi evolves, more and more projects are focusing on capital efficiency. I think somebody on Twitter made this observation that the propensity of all systems is to trend towards capital efficiency. And we're kind of seeing that in the Uniswap, v2 to v3 transition, and at the far end of that spectrum, the most capital efficient you can get is handling prices off-chain, because your liquidity is being allocated in that moment bespoke for the quote.
“...we're noticing is that as DeFi evolves, more and more projects are focusing on capital efficiency”
And so when people hear half a million in capital, that seems quite small compared to the billions of dollars that AMMs start measuring themselves by. But if you can use all of that capital very, very effectively, and you don't have billions of dollars sitting at the edges of that zero to infinity bonding curve, then you're getting the same result without letting a bunch of capital just sit there stagnant not gaining yield. So it's important to understand how these systems are maturing. And I think part of that dialogue in the DeFi space is lost in this memefied metric that we all go after instinctively.
CR: Makes sense. And so with this half a million in inventory, have you been able to get competitive quotes relative to Uniswap or Sushiswap or not yet? Do you think you still need more capital to get there?
VK: No, no, I think the quotes are pretty amazing at the moment. Alpha is live at the moment, so anybody who wants to check it out, they can go and sign up and we'll send you a link to access the alpha. You can see the prices yourself. So I don't know if I'm allowed to do a screen share on the podcast?
CR: Yeah, you can, people watching on YouTube will be able to see it and I'll narrate for people listening. So let's do a quick screen share. So Varun will share his screen now and show us what Hashflow can do.
VK: Or it might be easier to let people discover rather than me sharing the video, then I'm killing the suspense on the example.
CR: Right.
VK: Yeah, let's just do that, if that's the case. Let's just have people go and visit the website. I’m completely, pulling my shameless marketing plug here.
CR: That’s fine. Okay. So you have found that you're able to get good prices with that amount of liquidity and I did test it. You did send over the link and everything. And it was pretty good. They have this timer where they give you the latest quote, and it compares with Uniswap and Sushiswap. And Hashflow was getting better prices most of the time. And so for the times that the quote isn’t better to Uniswap, does the trade go through Uniswap?
VK: Yeah, so essentially, what we're saying is we're going to give you a better price than the best DEXs out there. But for whatever reason, if we cannot match the price, then we’ll execute on those exchanges. So if we can't beat the Uniswap price, then we'll execute on Uniswap.
PM: This comes back to that always on property that we talked about in terms of tradeoffs. 99% of the time, market makers will be quoting prices, because 99% of the time, the market makes sense. But when we have something like Black Thursday, March 12th when everything fell off a cliff, AMMs keep quoting in that environment. LPs basically have to keep selling their assets at fire sales even when the market is experiencing historic volatility. No sane market maker is going to do that, sell you their assets for very cheap, just because they have to or because you asked for it.
“...when we have something like Black Thursday, March 12th when everything fell off a cliff, AMMs keep quoting in that environment. LPs basically have to keep selling their assets at fire sales even when the market is experiencing historic volatility. No sane market maker is going to do that…”
So in those adverse situations, market makers might need time to recalibrate their strategy and pause sending quotes. But the beauty is because we have this base layer of DeFi that we've built over DeFi summer in the past year, that serves as a backstop to anything. So if you really need to sell your assets, we’ll route you to Uniswap, or Sushiswap, or whoever's giving the best price. But there's also this layer on top that is going to give you better prices, most of the time with the same trustless interface that you've come to expect.
CR: Got it. Okay, so you talk about this tradeoff that means that prices aren't always on. It also seems like you're centralizing part of the DEX on market makers. What are some of the risks of some Black Swan scenario that users should be aware of?
VK: I would say mostly, from the user standpoint, they're not giving custody to anybody, they're trading directly from the wallet to the pool. And if they don't like the quote, they can decide not to take the quote. So they don't really have that risk from a user standpoint. And from a market maker point of view, they could go and trade against themselves, but it doesn't really affect anybody else in the system. Especially because they own the pool separately, so from a trading point of view, the risks that a trader takes is maybe the same risk that anybody will be taking on other AMMs.
Some interesting things that came about were obviously the arbitrage opportunities that exist between the quote expiry, and the quote issuance and that could be much, much shorter than, say, 20 seconds, maybe 5 seconds or 10 seconds to prevent users from sitting on the quote and arbing the market maker. But I think that's again, the same risk they're taking on any DeFi which is a function of the constraints around Layer 1, on how long it takes for one block to be mined. So you can sit longer on a quote, but then you're also risking the quote might be mined at all, and the transaction does expire.
So that's some of the top level of attacks I can think of. Where the risk is more slightly geared towards the market makers side depending on if they are honest or not being honest. But then either way, it doesn't impact the trader, unless I missed something, Pranay?
PM: No, I think you got it. I think it's two sets of tradeoffs. One is around availability, and the other is fundamentally around the properties of the L1 that the quotes are being issued on. So from an availability standpoint, we already covered that where Uniswap’s price quotes are always available, but that it'll swing randomly on that curve, whereas market makers may cease quoting if they feel that the environment is not suitable. So that's the availability issue from a user standpoint.
The other tradeoff is around price impact. So with Uniswap, or Sushiswap, for the most part, you can get a quote from Uniswap and you can sit on it. The thing is, that price may move before you submit it on chain. I've had this experience where I put a low gas fee, and I send my unique transaction, and then when it actually gets mined on-chain, the transaction reverts because the slippage exceeded my limit. So that's not possible on Hashflow, because that quote is made for you, nobody can front-run it. The curve can't change underneath you and the transaction won't be reverted. But the tradeoff there of having this fixed quote that you can always take, is that quote will only be available for a certain amount of time. Because otherwise, if the market maker issued it for an hour, the market may drastically change and then that quote they offered you is not in their interest.
“I've had this experience where I put a low gas fee, and I send my unique transaction, and then when it actually gets mined on-chain, the transaction reverts because the slippage exceeded my limit. So that's not possible on Hashflow, because that quote is made for you, nobody can front-run it.”
Say they gave you a quote for ETH at, what is it 2,600 today, and then it goes up to 2,700 or something like that, now, they've just lost 100 bucks essentially. And so the market maker needs to put an expiry time on that quote to know that you will send it before the market changes too much. And so that tradeoff is handed to the user in the sense that the user needs to submit this quote in that short window. But for most use cases, a normal user should be submitting that transaction in MetaMask, or whatever wallet provider when they receive it. It's only the users that are kind of squatting on quotes or waiting for a while that may have an adverse experience. But I think the benefit of MEV resistance and avoiding slippage are quite powerful compared to this trivial ask of sending your transaction a little bit faster.
Waiting For a Mature Layer 2
CR: Makes sense. Cool. Why did you decide to build on Layer 1? Or are you thinking of going to Layer 2s gradually, or what's the thinking there?
PM: You want to take this one, Varun?
VK: Sure. In general, I think I'm waiting for a good Layer 2, so that's the first one. So at the moment, we just don't think L2s are mature enough for us to build on L2s just yet. And I think it will take at least maybe I could be wrong here, so maybe there'll be an L2 like next week, which is great if that happens. But I think we’re at least a year away from having a good L2 that addresses not just the technical tradeoffs, but also from a UX standpoint that is just as simple as using L1 Uniswap. And if there's any compromise of the UX side for the end user, then we wouldn't build on it just yet.
“I think we’re at least a year away from having a good L2 that addresses not just the technical tradeoffs, but also from a UX standpoint that is just as simple as using L1 Uniswap. And if there's any compromise of the UX side for the end user, then we wouldn't build on it just yet.”
So our focus as a product design focus has been to build the best product for the user. And for any reason if that UX has been compromised, then we wouldn't do it. Which is also why we're waiting for good L2 that also makes sense from a risk standpoint that actually is catering to the user and makes it really, really simple. And if for any reason, that doesn't work, then then we won't use it. So that's been my philosophy here. But Pranay, if you want to add something to that...
PM: I would just add that it's not a knock on L2s, l think we're all rooting for Ethereum’s success, right? I think Ethereum has shown itself to have the most network effects in terms of liquidity, developer interest and community in general. I think there's obviously a lot of other promising L1s and L2s as well. But pragmatically speaking, there's a tendency in the crypto space to hype something up at the very beginning because it shows a lot of promise. We saw this with Plasma. I think we're starting to see this with L2.
“...there's a tendency in the crypto space to hype something up at the very beginning because it shows a lot of promise. We saw this with Plasma. I think we're starting to see this with L2.”
The Optimism team is incredible and they're doing a lot of really amazing work with the OVM. But even they put out a blog recently called “optimistically cautious,” because I think they're trying to caution people that L2s have promise, but there is a maturation curve that needs to be scaled in the next 6-12 months. And a lot of that is not only deploying that raw L2 and having that essential execution environment be available. But it's migrating liquidity. It's making sure that the wallets and the user interfaces are up to snuff such that all the experiences people can have on L1, you can get the same thing on L2.
And so I think if we are realistic about where all the action is, it is still on L1 right now. It's where you can take a flash loan on Uniswap and then do something else with another DeFi lego. It's going to take a long time in crypto time for that critical mass to migrate over to an L2. And so while we're waiting, why not take advantage of much better gas prices on L1? People want to trade where they interface with all their other products that they like and right now, that is still L1. We don't see that changing in the next three to six months at the very least.
CR: Okay. I have been surprised at all the activity on Polygon, especially with Aave recently, they were able to attract a huge amount of volume in a short period of time to Polygon. So I don't know your thoughts on that. It seems like there's a lot of momentum going to that specific L2. I’m wondering if it was because of the UX that you didn't think it was ready for primetime?
PM: It's not just the UX. Polygon is a fantastic idea. It's the internet of blockchains on Ethereum. And so you're moving to this model of app specific chains that are all able to inter-op with each other. It's very much built in the model of a Cosmos or Polkadot, if you want to consider the shared security. But at the end of the day, a chain is a chain is a chain. If everybody takes everything on L1 and puts it on one L2, it's going to have the same problems as it had on L1. You're going to have throughput bottlenecks, because everybody's trying to send transactions on the same settlement layer.
“...at the end of the day, a chain is a chain is a chain. If everybody takes everything on L1 and puts it on one L2, it's going to have the same problems as it had on L1. You're going to have throughput bottlenecks, because everybody's trying to send transactions on the same settlement layer.”
So then you have to have these app specific L2s where you have Aave on one Polygon L2, and then you have another DeFi project on another, you need to get them to interopt perfectly like you can on L1 right now. I haven't used Polygon outside of trading on Polymarket, so I don't know what that interop experience is like. But I do know that Polygon’s switch from proof of authority to proof of stake, I think still has a lot of work to be done to make sure that there's things like robust slashing and be tested out at scale.
And I think it offers a promising alternative if you're priced out of L1, you have this L2 environment you can transact in. But I think it's not yet ready to take all of the DeFi composability and drop it on one Polygon chain right away.
CR: Interesting.
PM: And that's a caveat that I haven't played around with Polygon all that much. I've heard great things. But just fundamentally speaking about the tech, there's no magic scaling solution that will allow you to take everything on Layer 1, drop it on one Layer 2, and call it a day.
“ But just fundamentally speaking about the tech, there's no magic scaling solution that will allow you to take everything on Layer 1, drop it on one Layer 2, and call it a day.”
Token Specifics
CR: Got it. Okay, super interesting. And you're launching this week, will you have a token?
VK: We will have a token and more token use to follow.
CR: Is the token coming out at the same time as mainnet?
VK: Yeah, it's actually on mainnet right now. But it’s a phased launch. So we did our internal testing for the last couple of weeks before pushing it out and making an announcement. And the way we want to do it, we want to keep this in alpha for the next couple of weeks and test the system for, of course, random users as opposed to a more controlled set of people trying to trade and see how the system behaves and then move towards a more general announcement.
We also anticipate using this time and momentum to build out the community and then create more excitement before we announce tokens. I think it makes sense to launch the product first and then have people experience Hashflow as a product before selling tokens. So in terms of token launch and the date, I think it's really, really subjective depending on how quickly we can get the product going. So right now, it's live in alpha on mainnet, so you can actually go and trade and it works. So the question is, how quickly can we move to general announcements and get tokens out.
CR: Got it, okay. And will the token be a governance token? Are you wanting to decentralize Hashflow, turning into a DAO?
VK: Yes. So I think fundamentally, right now we are using tokens for a, governance and b, potentially have people get access to a portion of the network fee depending on the ownership of the token. So very much like what Sushi and Uni have been doing, the standard playbook. On top of that, also allow people to use governance to decide, perhaps the total supply and maybe whether to add inflation or not to add inflation and things like that. Decisions could be subject to governance. But at the same time, I'm also a fan of keeping governance minimal if possible, and mostly use tokens for other network utilities. But I think that's something we plan to determine in the upcoming weeks and release the specifics.
Permissionless Listing
CR: Cool. Oh, something I forgot to ask before, is listing tokens on Hashflow as decentralized as it is on other DEXs?
VK: Pranay, you…
PM: Yeah, yeah, I think what you’re getting at Cami is one of the benefits of Uniswap has been this permissionless listing quality that anybody can come in and drop an asset pair, and basically allow people to start trading. But we have seen some issues with that, namely, that price discovery is kind of challenging. At the beginning, you kind of have sandwich attacks and people front running, there are bots.
There's this cat and mouse game going on that I've been following on Twitter, where there are bots that try to scoop a new token listing, and then there are now bots that attack the bots. Like I think there's one called Salmonella that tries to find bots that are trying to find tokens that are newly listed, and it snipes the bots that are taking advantage of front running opportunities.
So there's still the raw advantage. I think the benefit of listing a token potentially on Hashflow is that the asset issuer or the lister could actually be the market maker themselves. So you can create, you can spin up a liquidity pool, capitalize it with your own inventory of the token that you minted, and then start offering the prices at whatever rate that you want. In this way, it can even be like a token offering conducted through Hashflow.
“...the benefit of listing a token potentially on Hashflow is that the asset issuer or the lister could actually be the market maker themselves. So you can create, you can spin up a liquidity pool, capitalize it with your own inventory of the token that you minted, and then start offering the prices at whatever rate that you want. In this way, it can even be like a token offering conducted through Hashflow.”
And the benefit of this is you avoid all those MEV and front running attacks that we talked about, but you also have a lot more capital efficiency and control over your pricing structure. You're not just putting your tokens into a pool and then seeing what happens. You're doing a controlled sale through your own pool that you control and you can then, once price discovery starts maturing, you can then expand out to the other AMMs.
The other benefit is that there are what are called DMMs, Designated Market Makers that help new assets gain liquidity on centralized exchanges. And now if an asset issuer wants, they can use a DMM to market make on DeFi itself. So again, the traditional playbook that we had in centralized finance is now available on DeFi because, again, we go back to exploring that design space between centralized and decentralized. So the power of bringing market makers into DeFi is exhibited on this front as well.
Long Term Vision
CR: Got it, super interesting. Okay. And then we're running out of time, but to wrap up. I want to know your long term, big vision for Hashflow. If you're wildly successful 10 years, 20 years from now, where are you, what is Hashflow at this point?
PM: I think that's you, Varun.
VK: That’s me? Okay. So, that'd be great if that happens. Obviously we're starting off, and then gradually building and seeing where the product goes. In the current market, I think that the biggest need right now is to bring more price efficiency or capital efficiency into DeFi, and bring price discovery to DeFi and allow anyone to still be able to take advantage of permissionlessness while expanding this to a broader spectrum. But if this works, then we are looking at even more than TradFi, like the Hudson Rivers of the world to become part of this.
So if that cycle continues, where if you said Bitcoin was disrupting L1 money, then Ethereum brought in the whole disruption of Wall Street and making Wall Street accessible to all Street. So now you're saying is, can we build the world’s settlement layer, where we can actually process all the liquidity going through this? So it's like the new cryptographic DTCC, for lack of a better example. And maybe I should go back to explain what DTCC is, for the audience.
So in general, I think the way I'd put it is, I could see all the world's liquidity flowing through it. That's the grand ambitious vision. But the certain context of what DTCC is, is in traditional finance, you've got your New York Stock Exchange, the NASDAQ, etc, where people trade and then you've got something called DTC, which is the custodian of your assets, your stocks, etc, and then you have the clearing house, and then you have broker dealers who are members of DTC that basically keep the assets in custody with DTC and then they go and trade on these exchanges.
And then periodically at the end of the day trade, after 5pm in New York time, they would go and do a net settlement of their final balances, and the cycle would continue. But your DTC acts as its custodian and the settlement layer for all the transactions that happen in traditional stock markets. So now what we're doing is we're basically taking all that complexity of humans and middlemen, and the lifecycle of an asset all the way from issuance, all the way to settlement, and physical delivery, we're taking all that complex logic and putting it into a thousand lines of code. And we are essentially making that system efficient by a thousand fold.
“So now what we're doing is we're basically taking all that complexity of humans and middlemen, and the lifecycle of an asset all the way from issuance, all the way to settlement, and physical delivery, we're taking all that complex logic and putting it into a thousand lines of code.”
So now, once they get the value of this, I think the path to where I said, all the world's liquidity will flow through this, is not as ambitious it sounds like. The question is how do you stage it, how do you go towards that? So you're starting with unbundling Uniswap, rather. But the sky's the limit or maybe the limit is beyond the sky.
CR: Very cool. Yeah, for those of us in DeFi, and seeing it every day, I don't think it's so farfetched to think that the world's liquidity will flow through the system, whether it's Hashflow, or some other DEX or a combination of many DEXs, eventually, it's just a better, more efficient system. Pranay do you want to add your big picture for both Hashflow and DeFi?
PM: Sure, yeah. For me, to come full circle, it's about sovereignty, it's preserving our rights in the digital frontier that, as we learned over the past year, spend most of our lives in now. And so a big step of that is finance, it's the energy of currency that we use to transact with each other. And what we're seeing is we saw that institutional adoption of Bitcoin kind of really take off this past year. People are not taking it as a joke, they're taking it seriously.
“For me, to come full circle, it's about sovereignty, it's preserving our rights in the digital frontier that, as we learned over the past year, spend most of our lives in now. And so a big step of that is finance, it's the energy of currency that we use to transact with each other.”
And that same thing is going to happen to DeFi. We're in the early stages where we're proving that toys are a precursor to the serious things phase of DeFi. But we're building a lot of the mechanics and the systems that will lay a foundation to take root, if not on Ethereum as Wall Street of crypto, but other Layer 1s and 2s, like we talked about earlier.
So part of that is bringing the old garden. If we want to have a fair world, if we want to create financial systems that give people access and don't leave them stranded the way they are in the current financial system, we do need to change but it doesn't mean throwing the baby out with the bathwater. We don't just get rid of all that expertise and all the 500-1000s of years of expertise that we've refined in traditional finance. We need to bring that over to DeFi but in a way that allows them to play this new game that we've all created.
“If we want to have a fair world, if we want to create financial systems that give people access and don't leave them stranded the way they are in the current financial system, we do need to change but it doesn't mean throwing the baby out with the bathwater. We don't just get rid of all that expertise and all the 500-1000s of years of expertise that we've refined in traditional finance.”
And so I see the first step as having the market makers being able to price assets on chain. But beyond that, who knows? Once that institutional liquidity, once that trickle comes in, the gates are going to open, it's going to be a flood and it's not going to just flood Ethereum, it's going to flood other L1s. And the beautiful thing about building apps is you can deploy everywhere. And so the goal is to be that bridge for all that liquidity, all that excellence, all that expertise that exists in the traditional world. Let's get that into DeFi and all of us who have been here from the early days will see the stuff that we're working on manifest into reality.
“Once that institutional liquidity, once that trickle comes in, the gates are going to open, it's going to be a flood and it's not going to just flood Ethereum, it's going to flood other L1s. And the beautiful thing about building apps is you can deploy everywhere.”
CR: Will be interesting to see what happens when institutions come to DeFi. I think things will get pretty crazy when that happens. They're already pretty wild right now. So…
PM: Yeah, I was going to ask it's pretty crazy now right?
CR: Yeah. So it’ll be a fun thing to cover in The Defiant. Anyways guys, what an interesting conversation, really a pleasure chatting with you. Thank you so much for joining. And we'll be keeping track of Hashflow and obviously, DEXs and DeFi so everyone subscribe to the newsletter, keep watching the YouTube channel and listening to the podcast. Thanks so much, Varun and Pranay.
The Defiant is a daily newsletter focusing on decentralized finance, a new financial system that’s being built on top of open blockchains. The space is evolving at breakneck speed and revolutionizing tech and money. Spread the word and share!
